Podcasts about Skylake

Play Episode Listen Later Sep 20, 2018 71:58

FreeBSD and DragonflyBSD benchmarks on AMD’s Threadripper, NetBSD 7.2 has been released, optimized out DTrace kernel symbols, stuck UEFI bootloaders, why ed is not a good editor today, tell your BSD story, and more. ##Headlines FreeBSD & DragonFlyBSD Put Up A Strong Fight On AMD’s Threadripper 2990WX, Benchmarks Against Linux The past two weeks I have been delivering a great deal of AMD Threadripper 2990WX benchmarks on Linux as well as some against Windows and Windows Server. But recently I got around to trying out some of the BSD operating systems on this 32-core / 64-thread processor to see how they would run and to see whether they would have similar scaling issues or not like we’ve seen on the Windows side against Linux. In this article are FreeBSD and DragonFlyBSD benchmarks with the X399 + 2990WX compared to a few Linux distributions. The BSDs I focused my testing on were FreeBSD 11.2-STABLE and 12.0-CURRENT/ALPHA1 (the version in development) as well as iX System’s TrueOS that is tracking FreeBSD 12.0-CURRENT. Also included were DragonFlyBSD, with FreeBSD and DragonFlyBSD being tied as my favorite operating systems when it comes to the BSDs. When it came to FreeBSD 11.2-STABLE and 12.0-ALPHA1 on the Threadripper 2990WX, it worked out surprisingly well. I encountered no real issues during my two days of benchmarking on FreeBSD (and TrueOS). It was a great experience and FreeBSD was happy to exploit the 64 threads on the system. DragonFlyBSD was a bit of a different story… Last week when I started this BSD testing I tried DragonFly 5.2.2 as the latest stable release as well as a DragonFlyBSD 5.3 development snapshot from last week: both failed to boot in either BIOS or UEFI modes. But then a few days ago DragonFlyBSD lead developer Matthew Dillon bought himself a 2990WX platform. He made the necessary changes to get DragonFlyBSD 5.3 working and he ended up finding really great performance and potential out of the platform. So I tried the latest DragonFlyBSD 5.3 daily ISO on 22 August and indeed it now booted successfully and we were off to the races. Thus there are some DragonFlyBSD 5.3 benchmarks included in this article too. Just hours ago, Matthew Dillon landed some 2990WX topology and scheduler enhancements but that fell out of the scope of when DragonFly was installed on this system. But over the weekend or so I plan to re-test DragonFlyBSD 5.3 and see how those optimizations affect the overall 2990WX performance now on that BSD. DragonFlyBSD 5.4 stable should certainly be an interesting release on several fronts! With FreeBSD 11.2-STABLE and 12.0-ALPHA1 I ran benchmarks when using their stock compiler (LLVM Clang 6.0) as well as GCC 7.3 obtained via GCC 7.3. That was done to rule out compiler differences in benchmarking against the GCC-based Linux distributions. On DragonFlyBSD 5.3 it defaults to the GCC 5.4.1 but via pkg I also did a secondary run when upgraded to GCC 7.3. The hardware and BIOS/UEFI settings were maintained the same throughout the entire benchmarking process. The system was made up of the AMD Ryzen Threadripper 2990WX at stock speeds, the ASUS ROG ZENITH EXTREME motherboard, 4 x 8GB DDR4-3200MHz memory, Samsung 970 EVO 500GB NVMe SSD, and Radeon RX Vega 56 graphics card. All of these Linux vs. BSD benchmarks were carried out in a fully-automated and reproducible manner using the open-source Phoronix Test Suite benchmarking framework. While for the last of today’s BSD vs. Linux benchmarking on the Threadripper 2990WX, the Linux distributions came out slightly ahead of FreeBSD and DragonFlyBSD with GCC (another test having issues with Clang 6.0 on the BSDs). Overall, I was quite taken away by the BSD performance on the Threadripper 2990WX – particularly FreeBSD. In a surprising number of benchmarks, the BSDs were outperforming the tested Linux distributions though often by incredibly thin margins. Still, quite an accomplishment for these BSD operating systems and considering how much better Linux is already doing than Windows 10 / Windows Server on this 32-core / 64-thread processor. Then again, the BSDs like Linux have a long history of running on high core/thread-count systems, super computers, and other HPC environments. It will be interesting to see how much faster DragonFlyBSD can run given today’s commit to its kernel with scheduler and topology improvements for the 2990WX. Those additional DragonFlyBSD benchmarks will be published in the coming days once they are completed. ###NetBSD 7.2 released The NetBSD Project is pleased to announce NetBSD 7.2, the second feature update of the NetBSD 7 release branch. It represents a selected subset of fixes deemed important for security or stability reasons, as well as new features and enhancements. General Security Note The NetBSD 7.2 release is a maintenance release of the netbsd-7 branch, which had it's first major release, NetBSD 7.0 in September 2015. A lot of security features have been added to later NetBSD versions, and for new installations we highly recommend using our latest release, NetBSD 8.0 instead. Some highlights of the 7.2 release are: Support for USB 3.0. Enhancements to the Linux emulation subsystem. Fixes in binary compatibility for ancient NetBSD executables. iwm(4) driver for Intel Wireless 726x, 316x, 826x and 416x series added. Support for Raspberry Pi 3 added. Fix interrupt setup on Hyper-V VMs with Legacy Network Adapter. SVR4 and IBCS2 compatibility subsystems have been disabled by default (besides IBCS2 on VAX). These subsystems also do not auto-load their modules any more. Various USB stability enhancements. Numerous bug fixes and stability improvements. Complete source and binaries for NetBSD 7.2 are available for download at many sites around the world. A list of download sites providing FTP, AnonCVS, SUP, and other services may be found at https://www.NetBSD.org/mirrors/. We encourage users who wish to install via ISO or USB disk images to download via BitTorrent by using the torrent files supplied in the images area. A list of hashes for the NetBSD 7.2 distribution has been signed with the well-connected PGP key for the NetBSD Security Officer: https://cdn.NetBSD.org/pub/NetBSD/security/hashes/NetBSD-7.2_hashes.asc NetBSD is free. All of the code is under non-restrictive licenses, and may be used without paying royalties to anyone. Free support services are available via our mailing lists and website. Commercial support is available from a variety of sources. More extensive information on NetBSD is available from our website: ##News Roundup Including optimized-out kernel symbols in dtrace on FreeBSD Have you ever had dtrace(1) on FreeBSD fail to list a probe that should exist in the kernel? This is because Clang will optimize-out some functions. The result is ctfconvert(1) will not generate debugging symbols that dtrace(1) uses to identify probes. I have a quick solution to getting those probes visible to dtrace(1). In my case, I was trying to instrument on ieee80211_ioctl_get80211, whose sister function ieee80211_ioctl_set80211 has a dtrace(1) probe in the generic FreeBSD 11 and 12 kernels. Both functions are located in /usr/src/sys/net80211/ieee80211_ioctl.c. My first attempt was to add to /etc/make.conf as follows and recompile the kernel. CFLAGS+=-O0 and -fno-inline-functions This failed to produce the dtrace(1) probe. Several other attempts failed and I was getting inconsistent compilation results (Is it me or is ieee80211_ioctl.c compiled with different flags if NO_CLEAN=1 is set?). When I manually compiled the object file by copying the compilation line for the object file and adding -O0 -fno-inline-functions, nm(1) on both the object file and kernel demonstrated that the symbol was present. I installed the kernel, rebooted and it was listed as a dtrace probe. Great! But as I continued to debug my WiFi driver (oh yeah, I’m very slowly extending rtwn(4)), I found myself rebuilding the kernel several times and frequently rebooting. Why not do this across the entire kernel? After hacking around, my solution was to modify the build scripts. My solution was to edit /usr/src/sys/conf/kern.pre.mk and modify all optimization level 2 to optimization level 0. The following is my diff(1) on FreeBSD 12.0-CURRENT. A few thoughts: This seems like a hack rather than a long-term solution. Either the problem is with the hard-coded optimization flags, or the inability to overwrite them in all places in make.conf. Removing optimizations is only something I would do in a non-production kernel, so its as if I have to choose between optimizations for a production kernel or having dtrace probes. But dtrace explicitly markets itself as not impactful on production. Using the dtrace pony as your featured image on WordPress does not render properly and must be rotated and modified. Blame Bryan Cantrill. If you have a better solution, please let me know and I will update the article, but this works for me! ###FreeBSD: UEFI Bootloader stuck on BootCurrent/BootOrder/BootInfo on Asus Motherboards (and fix!) Starting with FreeBSD CURRENT from about a few weeks of posting date, but including FreeBSD 12 alpha releases (not related to DEC Alpha), I noticed one thing: When I boot FreeBSD from UEFI on a homebuilt desktop with a Asus H87M-E motherboard, and have Root on ZFS, the bootloader gets stuck on lines like BootCurrent, BootOrder, and BootInfo. This issue occurs when I try to boot directly to efibootbootx64.efi. One person had a similar issue on a Asus H87I-PLUS motherboard. This issue may or may not exist on other Asus motherboards, desktops, or laptops. This may be specific to Asus motherboards for Intel’s Haswell, but may also exist on newer systems (e.g. Skylake) or older (e.g. Ivy Bridge) with Asus motherboards, as well as Asus desktops or laptops. There are two solutions to this problem: Use Legacy BIOS mode instead of UEFI mode Install a FreeBSD UEFI Boot entry Keep in mind that I am not going to talk about this issue and third-party UEFI boot managers such as rEFInd here. The first option is rather straightforward: you need to make sure your computer has “Secure Boot” disabled and “Legacy Boot” or “CSM” enabled. Then, you need to make sure FreeBSD is installed in BIOS mode. However, this solution is (in my opinion) suboptimal. Why? Because: You won’t be able to use hard drives bigger than 2TB You are limited to MBR Partitioning on Asus motherboards with UEFI as Asus motherboards refuse to boot GPT partitioned disks in BIOS mode Legacy BIOS mode may not exist on future computers or motherboards (although those systems may not have this issue, and this issue may get fixed by then) The second option, however, is less straightforward, but will let you keep UEFI. Many UEFI systems, including affected Asus motherboards described here, include a boot manager built into the UEFI. FreeBSD includes a tool called efibootmgr to manage this, similar to the similarly-named tool in Linux, but with a different syntax. ###Why ed(1) is not a good editor today I’ll start with my tweet: Heretical Unix opinion time: ed(1) may be the 'standard Unix editor', but it is not a particularly good editor outside of a limited environment that almost never applies today. There is a certain portion of Unixdom that really likes ed(1), the ‘standard Unix editor’. Having actually used ed for a not insignificant amount of time (although it was the friendlier ‘UofT ed’ variant), I have some reactions to what I feel is sometimes overzealous praise of it. One of these is what I tweeted. The fundamental limitation of ed is that it is what I call an indirect manipulation interface, in contrast to the explicit manipulation interfaces of screen editors like vi and graphical editors like sam (which are generally lumped together as ‘visual’ editors, so called because they actually show you the text you’re editing). When you edit text in ed, you have some problems that you don’t have in visual editors; you have to maintain in your head the context of what the text looks like (and where you are in it), you have to figure out how to address portions of that text in order to modify them, and finally you have to think about how your edit commands will change the context. Copious use of ed’s p command can help with the first problem, but nothing really deals with the other two. In order to use ed, you basically have to simulate parts of ed in your head. Ed is a great editor in situations where the editor explicitly presenting this context is a very expensive or outright impossible operation. Ed works great on real teletypes, for example, or over extremely slow links where you want to send and receive as little data as possible (and on real teletypes you have some amount of context in the form of an actual printout that you can look back at). Back in the old days of Unix, this described a fairly large number of situations; you had actual teletypes, you had slow dialup links (and later slow, high latency network links), and you had slow and heavily overloaded systems. However, that’s no longer the situation today (at least almost all of the time). Modern systems and links can easily support visual editors that continually show you the context of the text and generally let you more or less directly manipulate it (whether that is through cursoring around it or using a mouse). Such editors are easier and faster to use, and they leave you with more brainpower free to think about things like the program you’re writing (which is the important thing). If you can use a visual editor, ed is not a particularly good editor to use instead; you will probably spend a lot of effort (and some amount of time) on doing by hand something that the visual editor will do for you. If you are very practiced at ed, maybe this partly goes away, but I maintain that you are still working harder than you need to be. The people who say that ed is a quite powerful editor are correct; ed is quite capable (although sadly limited by only editing a single file). It’s just that it’s also a pain to use. (They’re also correct that ed is the foundation of many other things in Unix, including sed and vi. But that doesn’t mean that the best way to learn or understand those things is to learn and use ed.) This doesn’t make ed a useless, vestigial thing on modern Unix, though. There are uses for ed in non-interactive editing, for example. But on modern Unix, ed is a specialized tool, much like dc. It’s worth knowing that ed is there and roughly what it can do, but it’s probably not worth learning how to use it before you need it. And you’re unlikely to ever be in a situation where it’s the best choice for interactive editing (and if you are, something has generally gone wrong). (But if you enjoy exploring the obscure corners of Unix, sure, go for it. Learn dc too, because it’s interesting in its own way and, like ed, it’s one of those classical old Unix programs.) ##Beastie Bits Is there any interest in a #BSD user group in #Montreal? Tell your BSD story Finishing leftover tasks from Google Summer of Code Fuzzing the OpenBSD Kernel ARM - any Tier-1 *BSD options? ##Feedback/Questions Chris - byhve question Paulo - Topic suggestion Bostjan - How data gets to disk Send questions, comments, show ideas/topics, or stories you want mentioned on the show to feedback@bsdnow.tv

starting interview guide current modern code montreal root windows wifi intel tier samsung finishing wordpress arm fix how to usb stable tutorials gpt linux iso numerous amd install fixes enhancements news roundup raspberry pi bios dragonfly asus optimized trident vax ftp csm gcc unix hpc bittorrent pgp bsd windows server freebsd uoft copious clang uefi 2tb zfs secure boot openbsd threadripper fuzzing haswell google summer skylake refind alpha1 dtrace netbsd bsds matthew dillon ivy bridge trueos dragonflybsd llvm clang

Эпизод 94.1 - Slowly but surely (18.03.2018)

Securit13 Podcast

Play Episode Listen Later Mar 28, 2018 20:29

White House blasts Russia for NotPetya cyberattack https://edition.cnn.com/2018/02/15/politics/white-house-russia-notpetya/index.html Memcached servers can be hijacked for massive DDoS attacks https://www.networkworld.com/article/3258772/security/memcached-servers-can-be-hijacked-for-massive-ddos-attacks.html Memcrashed - Major amplification attacks from UDP port 11211 https://blog.cloudflare.com/memcr ashed-major-amplification-attacks-from-port-11211/ GITHUB SURVIVED THE BIGGEST DDOS ATTACK EVER RECORDED https://www.wired.com/story/github-ddos-memcached/amp NETSCOUT Arbor Confirms 1.7 Tbps DDoS Attack; The Terabit Attack Era Is Upon Us https://www.arbornetworks.com/blog/asert/netscout-arbor-confirms-1-7-tbps-ddos-attack-terabit-attack-era-upon-us/ У Харкові засуджено підозрюваного за продаж клієнтської бази поштового перевізника https://cyberpolice.gov.ua/news/u-xarkovi-zasudzheno-pidozryuvanogo-za-prodazh-kliyentskoyi-bazy-poshtovogo-pereviznyka-6604/ Speculative Execution Bounty Launch https://blogs.technet.microsoft.com/msrc/2018/03/14/speculative-execution-bounty-launch/ Frequently Asked Questions about Microsoft Bug Bounty Programs https://technet.microsoft.com/en-us/security/dn425055.aspx AMD allegedly has its own Spectre-like security flaws https://www.cnet.com/google-amp/news/amd-has-a-spectre-meltdown-like-security-flaw-of-its-own/ Linus Torvalds slams CTS Labs over AMD vulnerability report http://www.zdnet.com/article/linus-torvalds-slams-cts-labs-over-amd-vulnerability-report/ Intel: Our next chips won't have data leak flaws we told you totally not to worry about https://www.theregister.co.uk/2018/03/15/intel_spectre_mitigation/ Intel ships (hopefully stable) microcode for Skylake, Kaby Lake, Coffee Lake https://arstechnica.com/gadgets/2018/02/intel-ships-hopefully-stable-microcode-for-skylake-kaby-lake-coffee-lake/ Samba settings SNAFU lets any user change admin passwords https://www.theregister.co.uk/2018/03/14/samba_password_bug/ Zero-day vulnerability in Telegram https://securelist.com/zero-day-vulnerability-in-telegram/83800/ Plugins for Popular Text Editors Could Help Hackers Gain Elevated Privileges https://thehackernews.com/2018/03/text-editors-extensibility.html В Исландии похитили 600 серверов для добычи Bitcoin https://www.ixbt.com/news/2018/03/06/v-islandii-pohitili-600-serverov-dlja-dobychi-bitcoin.html CBM - Car Backdoor Maker https://www.kitploit.com/2018/03/cbm-car-backdoor-maker.html Let's Encrypt updates certificate automation, adds splats https://www.theregister.co.uk/2018/03/14/lets_encrypt_updates_certificate_automation_adds_splats/ CEO of smartmobe outfit Phantom Secure cuffed after cocaine sting, boast of murder-by-GPS http://www.theregister.co.uk/2018/03/13/phantom_secure_ceo_arrested/ Keygen Music [2+ hour Mix] https://www.youtube.com/watch?v=cYkaG5CT53I

229: The Meltdown of Spectre

Google Cloud Platform Podcast

Play Episode Listen Later Jan 17, 2018 108:40

We review Meltdown and Spectre responses from various BSD projects, show you how to run CentOS with bhyve, GhostBSD 11.1 is out, and we look at the case against the fork syscall. This episode was brought to you by Headlines More Meltdown Much has been happened this week, but before we get into a status update of the various mitigations on the other BSDs, some important updates: Intel has recalled the microcode update they issued on January 8th. It turns out this update can cause Haswell and Broadwell based systems to randomly reboot, with some frequency. (https://newsroom.intel.com/news/intel-security-issue-update-addressing-reboot-issues/) AMD has confirmed that its processors are vulnerable to both variants of Spectre, and the the fix for variant #2 will require a forthcoming microcode update, in addition to OS level mitigations (https://www.amd.com/en/corporate/speculative-execution) Fujitsu has provided a status report for most of its products, including SPARC hardware (https://sp.ts.fujitsu.com/dmsp/Publications/public/Intel-Side-Channel-Analysis-Method-Security-Review-CVE2017-5715-vulnerability-Fujitsu-products.pdf) The Register of course has some commentary (https://www.theregister.co.uk/2018/01/12/intel_warns_meltdown_spectre_fixes_make_broadwells_haswells_unstable/) If new code is needed, Intel will need to get it right: the company already faces numerous class action lawsuits. Data centre operators already scrambling to conduct unplanned maintenance will not be happy about the fix reducing stability. AMD has said that operating system patches alone will address the Spectre bounds check bypass bug. Fixing Spectre's branch target injection flaw will require firmware fixes that AMD has said will start to arrive for Ryzen and EPYC CPUs this week. The Register has also asked other server vendors how they're addressing the bugs. Oracle has patched its Linux, but has told us it has “No comment/statement on this as of now” in response to our query about its x86 systems, x86 cloud, Linux and Solaris on x86. The no comment regarding Linux is odd as fixes for Oracle Linux landed here (https://linux.oracle.com/errata/ELSA-2018-4006.html) on January 9th. SPARC-using Fujitsu, meanwhile, has published advice (PDF) revealing how it will address the twin bugs in its servers and PCs, and also saying its SPARC systems are “under investigation”. Response from OpenBSD: (https://undeadly.org/cgi?action=article;sid=20180106082238) 'Meltdown, aka "Dear Intel, you suck"' (https://marc.info/?t=151521438600001&r=1&w=2) Theo de Raadt's response to Meltdown (https://www.itwire.com/security/81338-handling-of-cpu-bug-disclosure-incredibly-bad-openbsd-s-de-raadt.html) That time in 2007 when Theo talked about how Intel x86 had major design problems in their chips (https://marc.info/?l=openbsd-misc&m=118296441702631&w=2) OpenBSD gets a Microcode updater (https://marc.info/?l=openbsd-cvs&m=151570987406841&w=2) Response from Dragonfly BSD: (http://lists.dragonflybsd.org/pipermail/users/2018-January/313758.html) The longer response in four commits One (http://lists.dragonflybsd.org/pipermail/commits/2018-January/627151.html) Two (http://lists.dragonflybsd.org/pipermail/commits/2018-January/627152.html) Three (http://lists.dragonflybsd.org/pipermail/commits/2018-January/627153.html) Four (http://lists.dragonflybsd.org/pipermail/commits/2018-January/627154.html) Even more Meltdown (https://www.dragonflydigest.com/2018/01/10/20718.html) DragonflyBSD master now has full IBRS and IBPB support (http://lists.dragonflybsd.org/pipermail/users/2018-January/335643.html) IBRS (Indirect Branch Restricted Speculation): The x86 IBRS feature requires corresponding microcode support. It mitigates the variant 2 vulnerability. If IBRS is set, near returns and near indirect jumps/calls will not allow their predicted target address to be controlled by code that executed in a less privileged prediction mode before the IBRS mode was last written with a value of 1 or on another logical processor so long as all RSB entries from the previous less privileged prediction mode are overwritten. Speculation on Skylake and later requires these patches ("dynamic IBRS") be used instead of retpoline. If you are very paranoid or you run on a CPU where IBRS=1 is cheaper, you may also want to run in "IBRS always" mode. IBPB (Indirect Branch Prediction Barrier): Setting of IBPB ensures that earlier code's behavior does not control later indirect branch predictions. It is used when context switching to new untrusted address space. Unlike IBRS, IBPB is a command MSR and does not retain its state. DragonFlyBSD's Meltdown Fix Causing More Slowdowns Than Linux (https://www.phoronix.com/scan.php?page=article&item=dragonfly-bsd-meltdown&num=1) NetBSD HOTPATCH() (http://mail-index.netbsd.org/source-changes/2018/01/07/msg090945.html) NetBSD SVS (Separate Virtual Space) (http://mail-index.netbsd.org/source-changes/2018/01/07/msg090952.html) Running CentOS with Bhyve (https://www.daemon-security.com/2018/01/bhyve-centos-0110.html) With the addition of UEFI in FreeBSD (since version 11), users of bhyve can use the UEFI boot loader instead of the grub2-bhyve port for booting operating systems such as Microsoft Windows, Linux and OpenBSD. The following page provides information necessary for setting up bhyve with UEFI boot loader support: https://wiki.freebsd.org/bhyve/UEFI Features have been added to vmrun.sh to make it easier to setup the UEFI boot loader, but the following is required to install the UEFI firmware pkg: # pkg install -y uefi-edk2-bhyve With graphical support, you can use a vnc client like tigervnc, which can be installed with the following command: # pkg install -y tigervnc In the case of most corporate or government environments, the Linux of choice is RHEL, or CentOS. Utilizing bhyve, you can test and install CentOS in a bhyve VM the same way you would deploy a Linux VM in production. The first step is to download the CentOS iso (for this tutorial I used the CentOS minimal ISO): http://isoredirect.centos.org/centos/7/isos/x8664/CentOS-7-x8664-Minimal-1708.iso I normally use a ZFS Volume (zvol) when running bhyve VMs. Run the following commands to create a zvol (ensure you have enough disk space to perform these operations): # zfs create -V20G -o volmode=dev zroot/centos0 (zroot in this case is the zpool I am using) Similar to my previous post about vmrun.sh, you need certain items to be configured on FreeBSD in order to use bhyve. The following commands are necessary to get things running: ``` echo "vfs.zfs.vol.mode=2" >> /boot/loader.conf kldload vmm ifconfig tap0 create sysctl net.link.tap.uponopen=1 net.link.tap.uponopen: 0 -> 1 ifconfig bridge0 create ifconfig bridge0 addm em0 addm tap0 ifconfig bridge0 up ``` (replace em0 with whatever your physical interface is). There are a number of utilities that can be used to manage bhyve VMs, and I am sure there is a way to use vmrun.sh to run Linux VMs, but since all of the HowTos for running Linux use the bhyve command line, the following script is what I use for running CentOS with bhyve. ``` !/bin/sh General bhyve install/run script for CentOS Based on scripts from pr1ntf and lattera HOST="127.0.0.1" PORT="5901" ISO="/tmp/centos.iso" VMNAME="centos" ZVOL="centos0" SERIAL="nmda0A" TAP="tap1" CPU="1" RAM="1024M" HEIGHT="800" WIDTH="600" if [ "$1" == "install" ]; then Kill it before starting it bhyvectl --destroy --vm=$VMNAME bhyve -c $CPU -m $RAM -H -P -A -s 0,hostbridge -s 2,virtio-net,$TAP -s 3,ahci-cd,$ISO -s 4,virtio-blk,/dev/zvol/zroot/$ZVOL -s 29,fbuf,tcp=$HOST:$PORT,w=$WIDTH,h=$HEIGHT -s 30,xhci,tablet -s 31,lpc -l com1,/dev/$SERIAL -l bootrom,/usr/local/share/uefi-firmware/BHYVE_UEFI.fd $VMNAME kill it after bhyvectl --destroy --vm=$VMNAME elif [ "$1" == "run" ]; then Kill it before starting it bhyvectl --destroy --vm=centos bhyve -c $CPU -m $RAM -w -H -s 0,hostbridge -s 2,virtio-net,$TAP -s 4,virtio-blk,/dev/zvol/zroot/$ZVOL -s 29,fbuf,tcp=$HOST:$PORT,w=$WIDTH,h=$HEIGHT -s 30,xhci,tablet -s 31,lpc -l com1,/dev/$SERIAL -l bootrom,/usr/local/share/uefi-firmware/BHYVE_UEFI.fd $VMNAME & else echo "Please type install or run"; fi ``` The variables at the top of the script can be adjusted to fit your own needs. With the addition of the graphics output protocol in UEFI (or UEFI-GOP), a VNC console is launched and hosted with the HOST and PORT setting. There is a password option available for the VNC service, but the connection should be treated as insecure. It is advised to only listen on localhost with the VNC console and tunnel into the host of the bhyve VM. Now with the ISO copied to /tmp/centos.iso, and the script saved as centos.sh you can run the following command to start the install: # ./centos.sh install At this point, using vncviewer (on the local machine, or over an SSH tunnel), you should be able to bring up the console and run the CentOS installer as normal. The absolutely most critical item is to resolve an issue with the booting of UEFI after the installation has completed. Because of the path used in bhyve, you need to run the following to be able to boot CentOS after the installation: # cp -f /mnt/sysimage/boot/efi/EFI/centos/grubx64.efi /mnt/sysimage/boot/efi/EFI/BOOT With this setting changed, the same script can be used to launch your CentOS VM as needed: # ./centos.sh run If you are interested in a better solution for managing your Linux VM, take a look at the various bhyve management ports in the FreeBSD ports tree. Interview - newnix architect - @newnix (https://bsd.network/@newnix) News Roundup GhostBSD 11.1 - FreeBSD for the desktop (https://distrowatch.com/weekly.php?issue=20180108#ghostbsd) GhostBSD is a desktop oriented operating system which is based on FreeBSD. The project takes the FreeBSD operating system and adds a desktop environment, some popular applications, a graphical package manager and Linux binary compatibility. GhostBSD is available in two flavours, MATE and Xfce, and is currently available for 64-bit x86 computers exclusively. I downloaded the MATE edition which is available as a 2.3GB ISO file. Installing GhostBSD's system installer is a graphical application which begins by asking us for our preferred language, which we can select from a list. We can then select our keyboard's layout and our time zone. When it comes to partitioning we have three main options: let GhostBSD take over the entire disk using UFS as the file system, create a custom UFS layout or take over the entire disk using ZFS as the file system. UFS is a classic file system and quite popular, it is more or less FreeBSD's equivalent to Linux's ext4. ZFS is a more advanced file system with snapshots, multi-disk volumes and optional deduplication of data. I decided to try the ZFS option. Once I selected ZFS I didn't have many more options to go through. I was given the chance to set the size of my swap space and choose whether to set up ZFS as a plain volume, with a mirrored disk for backup or in a RAID arrangement with multiple disks. I stayed with the plain, single disk arrangement. We are then asked to create a password for the root account and create a username and password for a regular user account. The installer lets us pick our account's shell with the default being fish, which seemed unusual. Other shells, including bash, csh, tcsh, ksh and zsh are available. The installer goes to work copying files and offers to reboot our computer when it is done. Early impressions The newly installed copy of GhostBSD boots to a graphical login screen where we can sign into the account we created during the install process. Signing into our account loads the MATE 1.18 desktop environment. I found MATE to be responsive and applications were quick to open. Early on I noticed odd window behaviour where windows would continue to slide around after I moved them with the mouse, as if the windows were skidding on ice. Turning off compositing in the MATE settings panel corrected this behaviour. I also found the desktop's default font (Montserrat Alternates) to be hard on my eyes as the font is thin and, for lack of a better term, bubbly. Fonts can be easily adjusted in the settings panel. A few minutes after I signed into my account, a notification appeared in the system tray letting me know software updates were available. Clicking the update icon brings up a small window showing us a list of package updates and, if any are available, updates to the base operating system. FreeBSD, and therefore GhostBSD, both separate the core operating system from the applications (packages) which run on the operating system. This means we can update the core of the system separately from the applications. GhostBSD's core remains relatively static and minimal while applications are updated using a semi-rolling schedule. When we are updating the core operating system, the update manager will give us the option of rebooting the system to finish the process. We can dismiss this prompt to continue working, but the wording of the prompt may be confusing. When asked if we want to reboot to continue the update process, the options presented to us are "Continue" or "Restart". The Continue option closes the update manager and returns us to the MATE desktop. The update manager worked well for me and the only issue I ran into was when I dismissed the update manager and then wanted to install updates later. There are two launchers for the update manager, one in MATE's System menu and one in the settings panel. Clicking either of these launchers didn't accomplish anything. Running the update manager from the command line simply caused the process to lock up until killed. I found if I had dismissed the update manager once, I'd have to wait until I logged in again to use it. Alternatively, I could use a command line tool or use the OctoPkg package manager to install package updates. Conclusions Most of my time with GhostBSD, I was impressed and happy with the operating system. GhostBSD builds on a solid, stable FreeBSD core. We benefit from FreeBSD's performance and its large collection of open source software packages. The MATE desktop was very responsive in my trial and the system is relatively light on memory, even when run on ZFS which has a reputation for taking up more memory than other file systems. FreeBSD Looks At Making Wayland Support Available By Default (https://www.phoronix.com/scan.php?page=news_item&px=FreeBSD-Wayland-Availability) There's an active discussion this week about making Wayland support available by default on FreeBSD. FreeBSD has working Wayland support -- well, assuming you have working Intel / Radeon graphics -- and do have Weston and some other Wayland components available via FreeBSD Ports. FreeBSD has offered working Wayland support that is "quite usable" for more than one year. But, it's not too easy to get going with Wayland on FreeBSD. Right now those FreeBSD desktop users wanting to use/develop with Wayland currently need to rebuild the GTK3 tool-kit, Mesa, and other packages with Wayland support enabled. This call for action now is about allowing the wayland=on to be made the default. This move would then allow these dependencies to be built with Wayland support by default, but for the foreseeable future FreeBSD will continue defaulting to X.Org-based sessions. The FreeBSD developers mostly acknowledge that Wayland is the future and the cost of enabling Wayland support by default is just slightly larger packages, but that weight is still leaner than the size of the X.Org code-base and its dependencies. FreeBSD vote thread (https://lists.freebsd.org/pipermail/freebsd-ports/2017-December/111906.html) TrueOS Fliped the switch already (https://github.com/trueos/trueos-core/commit/f48dba9d4e8cefc45d6f72336e7a0b5f42a2f6f1) fork is not my favorite syscall (https://sircmpwn.github.io/2018/01/02/The-case-against-fork.html) This article has been on my to-write list for a while now. In my opinion, fork is one of the most questionable design choices of Unix. I don't understand the circumstances that led to its creation, and I grieve over the legacy rationale that keeps it alive to this day. Let's set the scene. It's 1971 and you're a fly on the wall in Bell Labs, watching the first edition of Unix being designed for the PDP-11/20. This machine has a 16-bit address space with no more than 248 kilobytes of memory. They're discussing how they're going to support programs that spawn new programs, and someone has a brilliant idea. “What if we copied the entire address space of the program into a new process running from the same spot, then let them overwrite themselves with the new program?” This got a rousing laugh out of everyone present, then they moved on to a better design which would become immortalized in the most popular and influential operating system of all time. At least, that's the story I'd like to have been told. In actual fact, the laughter becomes consensus. There's an obvious problem with this approach: every time you want to execute a new program, the entire process space is copied and promptly discarded when the new program begins. Usually when I complain about fork, this the point when its supporters play the virtual memory card, pointing out that modern operating systems don't actually have to copy the whole address space. We'll get to that, but first — First Edition Unix does copy the whole process space, so this excuse wouldn't have held up at the time. By Fourth Edition Unix (the next one for which kernel sources survived), they had wisened up a bit, and started only copying segments when they faulted. This model leads to a number of problems. One is that the new process inherits all of the parent's process descriptors, so you have to close them all before you exec another process. However, unless you're manually keeping tabs on your open file descriptors, there is no way to know what file handles you must close! The hack that solves this is CLOEXEC, the first of many hacks that deal with fork's poor design choices. This file descriptors problem balloons a bit - consider for example if you want to set up a pipe. You have to establish a piped pair of file descriptors in the parent, then close every fd but the pipe in the child, then dup2 the pipe file descriptor over the (now recently closed) file descriptor 1. By this point you've probably had to do several non-trivial operations and utilize a handful of variables from the parent process space, which hopefully were on the stack so that we don't end up copying segments into the new process space anyway. These problems, however, pale in comparison to my number one complaint with the fork model. Fork is the direct cause of the stupidest component I've ever heard of in an operating system: the out-of-memory (aka OOM) killer. Say you have a process which is using half of the physical memory on your system, and wants to spawn a tiny program. Since fork “copies” the entire process, you might be inclined to think that this would make fork fail. But, on Linux and many other operating systems since, it does not fail! They agree that it's stupid to copy the entire process just to exec something else, but because fork is Important for Backwards Compatibility, they just fake it and reuse the same memory map (except read-only), then trap the faults and actually copy later. The hope is that the child will get on with it and exec before this happens. However, nothing prevents the child from doing something other than exec - it's free to use the memory space however it desires! This approach now leads to memory overcommittment - Linux has promised memory it does not have. As a result, when it really does run out of physical memory, Linux will just kill off processes until it has some memory back. Linux makes an awfully big fuss about “never breaking userspace” for a kernel that will lie about memory it doesn't have, then kill programs that try to use the back-alley memory they were given. That this nearly 50 year old crappy design choice has come to this astonishes me. Alas, I cannot rant forever without discussing the alternatives. There are better process models that have been developed since Unix! The first attempt I know of is BSD's vfork syscall, which is, in a nutshell, the same as fork but with severe limitations on what you do in the child process (i.e. nothing other than calling exec straight away). There are loads of problems with vfork. It only handles the most basic of use cases: you cannot set up a pipe, cannot set up a pty, and can't even close open file descriptors you inherited from the parent. Also, you couldn't really be sure of what variables you were and weren't editing or allowed to edit, considering the limitations of the C specification. Overall this syscall ended up being pretty useless. Another model is posixspawn, which is a hell of an interface. It's far too complicated for me to detail here, and in my opinion far too complicated to ever consider using in practice. Even if it could be understood by mortals, it's a really bad implementation of the spawn paradigm — it basically operates like fork backwards, and inherits many of the same flaws. You still have to deal with children inheriting your file descriptors, for example, only now you do it in the parent process. It's also straight-up impossible to make a genuine pipe with posixspawn. (Note: a reader corrected me - this is indeed possible via posixspawnfileactionsadddup2.) Let's talk about the good models - rfork and spawn (at least, if spawn is done right). rfork originated from plan9 and is a beautiful little coconut of a syscall, much like the rest of plan9. They also implement fork, but it's a special case of rfork. plan9 does not distinguish between processes and threads - all threads are processes and vice versa. However, new processes in plan9 are not the everything-must-go fuckfest of your typical fork call. Instead, you specify exactly what the child should get from you. You can choose to include (or not include) your memory space, file descriptors, environment, or a number of other things specific to plan9. There's a cool flag that makes it so you don't have to reap the process, too, which is nice because reaping children is another really stupid idea. It still has some problems, mainly around creating pipes without tremendous file descriptor fuckery, but it's basically as good as the fork model gets. Note: Linux offers this via the clone syscall now, but everyone just fork+execs anyway. The other model is the spawn model, which I prefer. This is the approach I took in my own kernel for KnightOS, and I think it's also used in NT (Microsoft's kernel). I don't really know much about NT, but I can tell you how it works in KnightOS. Basically, when you create a new process, it is kept in limbo until the parent consents to begin. You are given a handle with which you can configure the process - you can change its environment, load it up with file descriptors to your liking, and so on. When you're ready for it to begin, you give the go-ahead and it's off to the races. The spawn model has none of the flaws of fork. Both fork and exec can be useful at times, but spawning is much better for 90% of their use-cases. If I were to write a new kernel today, I'd probably take a leaf from plan9's book and find a happy medium between rfork and spawn, so you could use spawn to start new threads in your process space as well. To the brave OS designers of the future, ready to shrug off the weight of legacy: please reconsider fork. Enable ld.lld as bootstrap linker by default on amd64 (https://svnweb.freebsd.org/changeset/base/327783) Enable ld.lld as bootstrap linker by default on amd64 For some time we have been planning to migrate to LLVM's lld linker. Having a man page was the last blocking issue for using ld.lld to link the base system kernel + userland, now addressed by r327770. Link the kernel and userland libraries and binaries with ld.lld by default, for additional test coverage. This has been a long time in the making. On 2013-04-13 I submitted an upstream tracking issue in LLVM PR 23214: [META] Using LLD as FreeBSD's system linker. Since then 85 individual issues were identified, and submitted as dependencies. These have been addressed along with two and a half years of other lld development and improvement. I'd like to express deep gratitude to upstream lld developers Rui Ueyama, Rafael Espindola, George Rimar and Davide Italiano. They put in substantial effort in addressing the issues we found affecting FreeBSD/amd64. To revert to using ld.bfd as the bootstrap linker, in /etc/src.conf set WITHOUTLLDBOOTSTRAP=yes If you need to set this, please follow up with a PR or post to the freebsd-toolchain mailing list explaining how default WITHLLDBOOTSTRAP failed for your use case. Note that GNU ld.bfd is still installed as /usr/bin/ld, and will still be used for linking ports. ld.lld can be installed as /usr/bin/ld by setting in /etc/src.conf WITH_LLD_IS_LLD=yes A followup commit will set WITHLLDIS_LD by default, possibly after Clang/LLVM/lld 6.0 is merged to FreeBSD. Release notes: Yes Sponsored by: The FreeBSD Foundation Followup: https://www.mail-archive.com/svn-src-all@freebsd.org/msg155493.html *** Beastie Bits BSDCAN2017 Interview with Peter Hessler, Reyk Floeter, and Henning Brauer (https://undeadly.org/cgi?action=article;sid=20171229080944) video (https://www.youtube.com/watch?v=e-Xim3_rJns) DSBMD (https://freeshell.de/~mk/projects/dsbmd.html) ccc34 talk - May contain DTraces of FreeBSD (https://media.ccc.de/v/34c3-9196-may_contain_dtraces_of_freebsd) [scripts to run an OpenBSD mirror, rsync and verify])(https://github.com/bluhm/mirror-openbsd) Old School PC Fonts (https://int10h.org/oldschool-pc-fonts/readme/) Feedback/Questions David - Approach and Tools for Snapshots and Remote Replication (http://dpaste.com/33HKKEM#wrap) Brian - Help getting my FreeBSD systems talking across the city (http://dpaste.com/3QWFEYR#wrap) Malcolm - First BSD Meetup in Stockholm happened and it was great (http://dpaste.com/1Z9Y8H1) Brad - Update on TrueOS system (http://dpaste.com/3EC9RGG#wrap) ***

interview guide pr running data system turning tools built run os register tap oracle port stockholm developers intel ram similar serial signing utilizing meltdown mate speculation raid mesa storage how to open source restart tutorials alas linux minimal fork iso spectre pcs nt vm alternatively height amd enable publications cpu solaris snapshots microsoft windows unix fonts fujitsu width vms wayland ssh digital ocean bell labs pdp ryzen bsd gnu msr backwards compatibility sparc efi freebsd centos uefi ufs llvm vnc rhel zfs oom openbsd rsb xfce haswell broadwell peter hessler skylake raadt linux vms netbsd oracle linux microcode allan jude bsds ixsystems tarsnap ghostbsd trueos dragonflybsd

CMP333: How to Get the HPC Best-in-class Performance via Intel Xeon Skylake Processors and AWS C5 Instances with Ease of Deployment Through Rescale

AWS re:Invent 2017

Play Episode Listen Later Nov 30, 2017 45:31

HPC Cloud services built on the latest Intel architecture, Skylake Xeon processor, are now powering the C5 compute intensive instance at AWS and can serve as your next-generation HPC platform. Hear how customers are starting to consider hybrid strategies to increase productivity and lower their capital expenditure and maintenance costs. Also learn how to adapt this model to meet the increasing HPC and data analytics needs for your applications with the new technologies incorporated into the platform. Also find out how high performance computing via Rescale's cloud platform using Intel's latest technology seamlessly brings these advantages to HPC management and users. To access team management and throughput, all can benefit from cloud platform adoption as demand increases. Learn how customers are already benefiting. Richard Childress Racing (RCR) discusses its use of AWS C5 via the Rescale platform and how the combination is giving it an edge in the highly competitive field of motorsport. Session sponsored by Intel

performance intel ease aws deployment c5 instances processors hpc best in class intel xeon skylake rescale

Node.js with Myles Borins

Play Episode Listen Later Nov 29, 2017 32:06

Myles Borins talks with Mark and Francesc about Node.js from its history, how to contribute, the consensus-seeking governance, and why it's important to Google Cloud Platform. Node.js is an open-source, JavaScript runtime environment built on Chrome's V8 JavaScript engine, and Google is a Platinum Member of the Node.js Foundation. About Myles Borins Myles Borins is a developer, musician, artist, and maker he works for Google as a developer advocate serving the Node.js ecosystem he graduated with a Master of Music Science and Technology from c.c.r.m.a. a.k.a the center for computer research in music and acoustics Cool things of the week Reduced GPU prices on GCP and preemptible local SSDs blog Skylake processors now available in 7 regions blog New Episodes of Learn TensorFlow and Deep Learning, without a PhD: Modern Convolutional Neural Nets video Modern RNN Architectures video Deep Reinforcement Leanring video Interview Node.js site Node.js on Google Cloud Platform site docs Node School site Node.js with Justin Beckwith podcast App Engine site docs Cloud Functions site docs Kubernetes Engine site doc Introduction to Kubernetes: How to Deploy a Node.js Docker App site Socket.io site Question of the week How do you give public postmordems? Fearless shared postmortems - CRE life lessons blog Where can you find us next? Mark will be Montreal in December to speak at Montreal International Games Summit. Melanie will be at SOCML (Self-Organizing Conference on Machine Learning) end of this week and NIPS (Neural Information Processing Systems) in Long Beach next week.

google master technology foundation montreal fearless machine learning long beach chrome javascript cre deep learning ssd node deploy gcp nodejs socket francesc google cloud platform skylake music science v8 javascript platinum member

EP107 - Listener Questions and Amazon News

The Jason & Scot Show - E-Commerce And Retail News

Play Episode Listen Later Nov 11, 2017 73:55

EP107- Listener Questions Listener Questions Q1: Brands selling direct on their own site Shawn Cheng What do u think about a brand to run their own brand store, not through market place such as eBay, Amazon or Alibaba? Jamie Dooley Hi Scot & Jason: Have you heard of any brands seeing true success building a D2C E-Comm business through their own websites? (On a path to do 10-20% of sales and/or 8-9 figures in annual sales?). D2C seems like D2C was a big trend a year ago but I am not hearing about success stories where sales justified significant spend. Scott Silverman Do you think a new brand without an e-commerce site or digital presence could be built by selling on Amazon? Secondarily, should manufacturers selling on their own e-comm sites, shut them down and just sell via retailers? Q2 Mobile Conversion Gap Ari Nahmani Mobile conversion rate... retailers are getting more and more of their web traffic from mobile, but those users are half or a third as likely to convert. They don’t seem to be coming back on desktop. So what’s happening? We see across the board where YoY traffic is flat, revenue is down due to the device mix over-indexing on mobile YoY. How do we explain this behavior? Where are those users purchasing if ecommerce growth is up? I’m seeing his trend on several client sites and I recall one of your shows that this trend was discussed. Q3: Omni-Channel Fulfillment Alexandro Volakis order sourcing in an omnichannel network. how do you decide where its best to ship from? Q4: Singles Day in the US Julia Ptock How do you think about Singles Day (11.11.)? Are there some retailers who take part at this event in the USA? Amazon is focusing on Cyber Week (Cyber Monday & Black Friday) but doesn't show interest in Single Day. Do you have an idea why? Q5: Toys R Us impact on Holiday Promotions Melissa Burdick How will the bankruptcy of Toys R Us impact Amazon this Holiday? Is it going to be a bloodbath in pricing this Holiday with TRU stores cutting prices and Amazon price matching (and then closing stores shortly after holiday)? https://www.usatoday.com/.../toys-r-us-store.../683762001/ Amazon News Amazon opening permanent device shops inside of some Chicago WholeFoods stores. Jason visited one under construction. Amazon lowers prices on marketplace sellers products. It looks like primarily 3000 beauty SKU's at the moment. More Amazon Private Label: Furniture: Rivet - “stylish and versatile mid-century modern furniture and décor". Mid-century modern aesthetic, focus on small space solutions, 1 year warranty, prime exclusive Stone & Beam - Higher end price points. modern farmhouse aesthetic, 3 year warranty. Athlesure Rebel Canyon - Low price point, Men's and Woman's lounge clothing. Peak Velocity - Higher price point active wear (Under Armor competitor), prime exclusive GoodSport - Men's and Woman's moisture wicking apparel Baby Mama-Bear - Relaunched diapers Amazon Private Label articles: Bloomberg piece on Amazon sportswear w/ L2 Glossy covers Amazon Gift Guide (which focuses on exclusive Amazon products) http://jasonandscot.com Join your hosts Jason "Retailgeek" Goldberg, SVP Commerce & Content at Razorfish, and Scot Wingo, Founder and Executive Chairman of Channel Advisor as they discuss the latest news and trends in the world of e-commerce and digital shopper marketing. Don't forget to like our facebook page, and if you enjoyed this episode please write us a review on itunes. Episode 107 of the Jason & Scot show was recorded on Thursday November 9, 2017. New beta feature - Google Automated Transcription of the show Transcript Jason: [0:25] Welcome to the Jason and Scott show this is episode 107 being recorded on Thursday November 9th 2017 I'm your host Jason retailgeek Goldberg and as usual I'm here with your co-host Scot Wingo. Scot: [0:40] Hey Jason and welcome back Jason Scott show listeners Jason how you doing. Jason: [0:46] I'm doing terrific Scott if you like there's a number of exciting things I've been eager to talk to you about. Scot: [0:53] Let's talk about the Chicago weather first. Jason: [0:55] It's sad I feel like we've had a super mild winter so far but this week it turned cold so we've been in the forties and tomorrow it's, can a potentially drop it on the twenties in snow so I've had to visit a portion of my closet I haven't seen in awhile. [1:15] What about Raleigh is it beautiful and sunny still. Scot: [1:18] It wasn't beautiful but it is still like in the 50s so I enjoy hearing about your snow stories. Jason: [1:24] I am always happy to help you feel better about your life by hearing about mine. Scot: [1:29] Thanks man. Jason: [1:30] I would imagine that you're feeling some extra warmth because I feel like there's been some exciting Star Wars announcements later that keep you up warm and comfy. Scot: [1:39] Yes it is a great time to be a Star Wars fan we went through kind of. Salmon there for a long time and I know we have the prequel trilogy which is exciting and then we didn't know what the future would hold and then with the Disney acquisition the announcements coming Fast and Furious so we have first of all we have. 2 Star Wars movies just right in the pipeline right behind each other which is exciting as we record this 35 days until the last Jedi 34 if you go on the. Actual opening on the 14th 196 days for the Solo movie so excited about that little Star Wars story and then the big news is at the. Industrial, conference call at Bob Iger announce to Star Wars things so Ryan Johnson the guy thats directing Last Jedi they loved working with him so much they've given him his own Trilogy so it's going to be some, three part story in the Star Wars universe but not part of the normal Saga and then they're also. I'm sure you seen this but they're doing a streaming, thing at Disney that this is all the rage everyone's unbundling and now will pay, 8 times as much for all the content but anyway they're doing a Disney streaming Channel and they announced a Star Wars live show will be on that so. Lot of great new Star Wars content con. Jason: [2:56] Yeah yeah yeah the TV show super exciting I'm with you I've been annoyed by all this unbundling like I, I suspect you and I both had to buy the CBS subscription to get Star Trek I probably would have already had to give the Disney one due to my son so maybe like in that case I'm not as upset but. [3:16] And you and I both been walking around with her iPhone 10s for almost 2 weeks now so what's what's the verdict for you. Scot: [3:23] It's awesome it is a great phone the. [3:28] Notches not a big deal the face ID is really cool I really like it it's very handy to buy stuff it is a little unusual to Skylake pick up your phone and look at it, but you get used to it, I have an Android phone that is the same feel as iPhone and everyone will take it out of my pocket and look at it and then I feel really ridiculous cuz it doesn't look back at me. [3:50] Just kind of like what you think I'm an iPhone and then I would put my phone and it unlocks to the face ID thing is pretty cool how do you like yours. Jason: [4:00] I can agree on how you're looking your Android phone is how my son looks at everything that isn't like an echo like he expects All Electronics to be able to talk to him. [4:11] But I have been super happy with the 10 I'm with you everything has worked pretty smooth. Uniden it I have is without the button there's not an obvious way to feel with the right orientation of the phone at so I now fine I put out of my pocket upside down or backwards more than I. I used to. Scot: [4:33] Get a feel for the camera bump. Jason: [4:35] Yeah I have. Scot: [4:37] About should be on your right index finger. Jason: [4:39] That that I would do you answers part of me that doesn't want to like get a bunch of smudgy fingerprints on the camera now that I'm not OCD. Scot: [4:48] That's why I put the camera there. Jason: [4:50] Gotcha I'm think I did put the leather case on it the kid cancer got the phone a lot so I felt like I needed some protection and so now I what I feel for is the. [5:02] There's no weather in the bottom of the phone so that's how you can tell. [5:06] Yeah so that's worked out well and what's killing me is I do use an iPad a fair amount and the muscle memory now to go back and forth between I feel like I've gotten used to the all the gestures. [5:20] Answer that the button still being on my iPad is killing me so I feel like I'm going to have to get a new iPad when they come out just to have them all work the same. Scot: [5:28] Yeah I like the new gestures I don't like swipe down from upper right to get to the control panel because I frequently hold it. One handed and like that's pretty weird gesture to do single-handed. [5:41] So I kind of don't like that and I did hear there's a rumor of their members on the. Apple Hardware sides today or I read a rumor they're working on a high-end iPad that will have face ID in a bunch of other cool stuff and then there's a lot of rumors going around about a r Hardware that they're working on which is kind of interesting. Jason: [6:00] Yeah man you know in some ways. [6:04] There's not new AR Hardware in this device but this is the first device with the horsepower to Sephora support the the new AR software kit and so there are some cool. [6:16] New AR apps that you can run on this phone that you couldn't run on other phones which is pretty cool we make that make him up in one of the listener question. Scot: [6:24] Yep and that this is nursing one of these kind of these designers kind of tries to pontificate what the future looks like these things that this interesting observation that the sensors that are in the notch on the 10. Are really everything you need for a r glasses today they took essentially kind of the the look and feel of the 10 and they put that not kind of on bridge of glasses and then they, they made glasses come around them which is kind of an interesting I have never thought of it but it is kind of interesting. Because it's doing the face ID in all the pieces you need to do that is exactly what you need to turn that around and be able look out at the world with with ar glasses so. [7:03] That could be kind of part of what they're thinking about. Jason: [7:05] So an interesting thing in this comes into play in some in-store retail environments the Microsoft Kinect has been the the cheapest ubiquitous sort of. [7:17] 3D camera with. [7:19] Infrared distance measuring that was out there until there were tons of little Nitch applications that hacked some solution using that. [7:27] Does Microsoft Kinect camera so there's lots of these like 3D body scanners that you could potentially use for. [7:35] Uploading your avatar to your favorite e-commerce site or or made to order clothing or measuring rooms for furniture and all all these different things in there a bunch of in-store applications where are these. Is Microsoft Kinect camera got hacked in Microsoft just announced that they're discontinuing the Kinect and so that's going to go away and of course the. Everything that was in that big camera module is now you know there's a better version in this little notch on the on the iPhone and so I think we're going to start to see a bunch of. Interesting apps and use cases where they're essentially using an iPhone just for that that sensor array and in a bunch of. I fixed installations in which I think would be cool you know the bummer at the moment is there going to be hard to Source in there going to be expensive for a while. [8:27] And I sure wish they had them on the back as well cuz I feel like. [8:30] Retail uses it would be very handy to have that sense of Ray pointing out so that we could use it for some other uses. Scot: [8:38] Vehicle in the other exciting gadgets tree portal. Jason: [8:42] Well I know you got a new one that I'm eager to hear about when we'll get to that in a minute the other thing I'm just over over all excited for this week, is so we recording this Thursday night which is Friday morning in China so it's the day before singles day we'll talk about that like that would be an exciting event on the show just anyway but it's extra exciting because, our very first show was a recap of singles day which tells me that the next should we do will be our our anniversary show. Scot: [9:14] Yeah if you're coming in on two years who knew. Jason: [9:17] Exactly I do think you put up with me for that one. Scot: [9:20] It's been a struggle but I managed to figure it out. Jason: [9:25] I appreciate it I'm I can't speak for the listeners that II. Scot: [9:28] Speed of a source of it's been awhile since we did a question episode, way back to episode 96 actually so put put a call out this morning and we actually got a very strong response our listeners have a lot of questions, so let's jump into someone's no questions. [9:59] Questions first one is not really questions more of a statement it's from Natalie Bowman and she says hello. Jason: [10:07] Hey Natalie thanks for responding to the questions. Scot: [10:11] Hi Natalie okay and then next question is kind of interesting three people. Who's similar nuances of the same kind of a topic so the first flavor of this was from Shawn ching and. I said what do you think about a brand to run their own Brand store not to a Marketplace such as eBay Amazon or Alibaba. So you should have ran have their own e-commerce site is kind of the flavor there and then are a good friend of the show Jamie Dooley asked hey Scott Jason have you heard of any brand seeing true success Building address. Consumer e-commerce business through their own website so should you do it and then have we seen anyone that's had success. I in any kind of adds a Nuance you know it seem like it was a big trend of year ago but I'm not hearing about success stories and then another friend of the show Scott's ornament he said what do you think about it there's a new brand. [11:10] That doesn't have any Commerce site should they just start on Amazon or Marketplace like that. And then should manufacturers be selling all their own e-commerce sites are just shutting down and Salvia retailers so. [11:23] Interesting flavor kind of in this topic of Brands going direct that we've hit on. Probably at least every other week if not every week but what's your advice when a brand comes to you with with those flavored questions. Jason: [11:35] Yeah so I agree they're all related to Sean and Scott's to me are almost identical right like the two spins on the exact same thing which is. Do you need to have your own branded site in addition to selling on the marketplaces are you can you get by with just having a presence on the market places in my strong advice to anyone, it's trying to build a long-term sustainable company is I would I would definitely encourage you to have your own. Site in addition to you know whatever efforts make sense for you on Market places in. The reason I say that is a couple in most cases you're not going to do the kind of volume on your own site that you're going to do on the marketplaces I get it in so it you know it maybe. Your first effort might be your presence on the marketplaces and it may not be that appealing to invest a ton in your own site. But the thing is. We are all essentially digital sharecroppers on the the marketplaces platforms right like. [12:41] They can change their terms and conditions at any time they could be horribly under favorable the. We could intentionally or unintentionally run a fall of any of their their policies or. [12:56] He perceived to run afoul of them and get cut out of those Marketplace it there all kinds of bad things that can happen on the marketplaces. And it didn't given time you look at it and say hey yeah I know there were some Old Market places that change the rules all the time but Amazon has been much more consistent or. Alibaba is much more consistent or whatever the case is. [13:16] Overtime that can just shift so it's just really risky to have a hundred percent of your. Brand presence be on this site that you don't own and you don't control and that your landlord can essentially. Raise your lease and change that your terms or kick you out in anytime and so I do think it it absolutely makes sense to have your own destination on the web that you own and you absolutely can. And to the extent that your brand can drive any organic traffic that you can build any of your own falling. It just makes more sense and it's more profitable to send those customers to your own site instead of. To the marketplace you can avoid the take rate you can still in most cases for Phil through you know whatever fulfillment vehicle you're using on the marketplace. And you know as you get to know some of those customers into the relationship with his customers like this gets a little dicey but you know there certainly is a, percentage of your customers you can shift to be direct on your own site and you just you know on those marketplaces you're totally disintermediated from the customer, and so you did you know even if it's only a small percentage of your customer base, you want a direct relationship with some customers if only to get feedback to be able to understand what kind of content is selling and not selling and to build to run a B test and do all sorts of other things so for those reasons, I would say you absolutely have to invest in your own site, and I'll you know I'll give you a caveat that it maybe isn't the first investment you make are the biggest investment you make that that fair to you Scott. Scot: [14:54] Yeah and a lot of it depends where you're coming from too so you know there's there's many buckets of Brands these days and it's. It's becoming increasingly easy to create a new brand when we were growing up, new brands had this like huge hurdle to launch them yet to do a TV campaign and all this kind of stuff and now we're just seeing an explosion of brand so so I would use a framework where there's kind of Legacy Brands and then, new kind of born recently kind of Brands and. Do I cycle I see with newly-born Brands is if we take and there's kind of two segments there but let's just take. More Scrappy auction real ones they to Sean's Point Sean ass this year they. They start on marketplaces soap to marketplaces a great place to start a colony e-commerce training wheels because they have these incumbent. [15:51] Consumers already there so just like riding a bike with training wheels it's hard to steer pedal and balance so training wheels takes balance out of the equation starting a direct-to-consumer business you know it's hard to acquire customers. Gold products can get the products to the consumers and and all that so I Marketplace simply gives you the training wheels by giving you a customer sand, I wanting them to you to your pointer sharecropping of whatever whatever analogy is there and but you know to your point. Next product life cycle needs to be maybe start on Amazon you go multi Marketplace but the sooner you can kind of create your own presents on the internet where you can control the brand at the better, and then there's a lot of tools to Jamie's Point kind of weave that in the companies that have had a substantial kind of was caught materials over 10% of their sales on their website, there's a lot of tricks that utilize to do that and what are the simplest ones is offering something special to your website customers. You could think about special pricing but that actually kind of creates this. [17:04] The problem so that's usually not what brands do but usually special products so for example Under Armour I don't know if their website is out there. To talk about this publicly I don't know if their websites 10% of their sales I doubt it is cuz they have that huge wholesale component yeah. But you're one of the clever things they do on their site and it did I know is successful is if you're an atheist of their brand that's where they launched a lot of new stuff so that's kind of like the exclusive Channel 4 new stuff and then there's this waterfall maybe the new stuff there for. [17:38] A month in an account waterfalls into retail and then it waterfalls down into a Marketplace or something like that that's interesting kind of thing a lot people do exclusives on different channels the mattresses and why electronic guys are King. the digitally native vertical Brands what's interesting about kind of those is they start really with a website and then we seen many of them. Going to realize you can only get so far with that approach so it's almost kind of speaks to the store location is the right strategy so they are diversifying into offline a lot of them are exploring Marketplace is a lot of exploring Retail Partners those kinds of think so. So I think the best strategy is a balanced kind of from a risk and a channel perspective is to have a portfolio of channels and that includes having website. Jason: [18:24] Yeah tonight I would totally agree in like just to tell you more explicitly answer Jamie's question, I bet you hit it like all those digital need a vertical brands, all you know hit the eight or nine figures in in direct e-commerce sales so that's in a bona Bose ModCloth Warby Parker Casper, you know all of those guys certainly do it there's some some pretty big brands that we don't hear about as much for e-commerce but you know I think the surprise people when you see how big they are but like revolve clothing I think is a big one and then of course stitchfix which arguably started out as a multi, vendor retailer but but is Shifting to a branded, to be more of a man with her own products I mean obviously you know got got pretty darn near a billion dollars. Predominantly through their own website so I think I think Jamie's right that the hype, was there before and I absolutely don't think it's one of these things were you build it in your guaranteed success so you know I think, to to Jamie's point there was probably a light-year a couple years ago when I went. I love I just want to website you I called you know I'll be. [19:41] Entitled to these sort of eight or nine figure, run rates and you know we certainly seen a lot of people fail but they're absolutely have been and continue to be some successful sites sites in that space. Scot: [19:56] It was talking about just failure for some kind of like what brands do wrong the number one thing I see is, the brands of everyone loves this map pricing concept and many Brands Don't Force It so it will do is still set for website and all, yeah they're their there. I have my pricing so they adhere to that obviously because they believed in the policy and they're selling stuff directly there and then they will have absolutely no kind of understanding of consumer. Expectations around shipping cost and time so they'll have you know a $50 widget for $10 shipping and, you know you can upgrade to 3 days for $40 and you know the $10 shipping is the equivalent of USPS. Yo week. Week plus you 7 to 10 day type of delivery without tracking and then they're shocked when they don't sell a lot because you know why we put up this website and you know we have all this traffic and no one's buying things why is that. Did he have to have that kind of discussion about well you're the single. Most expensive place to buy your products on the Internet is your website your shipping and you know the cost or just way off base with what consumers want and then another funny one is a lot of Brands and you know. To go to the legal department and all the stuff and they end up like not doing the basics like user-generated reviews and things like that because. [21:27] How to get really wrapped around the axle like what if someone leaves a bad review should we go delete that or what should we do and you know should we should we go after them with a cease-and-desist letter and it's kind of funny to have this discussion because. There are people reviewing their products right there on Amazon but it just shows you're somebody's Legacy companies have such a hard time wrapping your head around this digital world there's some the things I see that happen all the time or these Brands really get off base with her web store. Jason: [21:56] Yeah I know told you I've had all those conversations. That you another there is just a lot of back of house stuff that people tend to overlook when they're you know I used to wholesale model and they're going direct to Consumer for the first time so it's your point like, they generally wolfley underestimate fulfillment know where they are selling charging too much for 4, crappy level of service but it's probably also a side job for the wholesale fulfillment guys instead of stuff probably sits in the warehouse for 5 days after you place the order, before it even gets into the the the shipper system, answer their those issues and then like you know there's a customer service guy and they that that guy quickly gets overwhelmed with calls so they're all those kinds of things and, if you survive the infant mortality like if you survived all those mistakes the next big steak mistake we see everyone making is that there's, every Branford, whatever the brand attributes are whatever Niche it's in how well it's known there's some certain amount of sort of organic traffic that's relatively easy for each branch require in for some Brands that's. Three significant amount of traffic for some Brands that's not very significant but there always is some threshold where if you do the fundamentals right you get to a certain level and it then you hit a wall and it suddenly becomes much harder to grow inside the. [23:26] The real test for the sustainable direct-to-consumer business is are, you know what once you get over that that first tranche of easy to get customers can you, still be profitable and successful in growing beyond that original based, or you just met being a cop out and you get stuck there or do you start spending way too much on customer acquisition and I think that's a mistake we see a lot so, so they're there are definitely lots of pitfalls and there's some good examples of of companies that have been able to steer clear of them. Scot: [23:59] Let's jump in turn next question and then we can kind of Go Lightning round on a couple days maybe we'll see. Jason: [24:08] Wait that wasn't Lightning Run. Scot: [24:09] How's Jason Scott lightning room, okay second question this is from RE nahmani he is the CEO of an israel-based digital agency, and he says I'd like to talk about mobile conversion rate retailers are getting more and more of their web traffic from mobile yes but those users are 1/2 or 1/3 is likely to convert, they don't seem to be coming back on desktop so what's happening we see this across-the-board we're year-over-year traffic flat revenues down due to the vice mix over indexing on mobile how do we think about this Behavior. Where are the users buying stuff e-commerce is growing let me see okay so yes. [24:53] And then I'll take that over to you cuz you have a clever name for it. Jason: [24:56] So we've talked about that on a couple episodes I call it the mobile Gap and it it's it's very real you know most sides are seeing, their Mobile Traffic grow much faster than their desktop traffic so they often would characterize that as, there traffic is shifting from desktop to mobile and the conversion rate on that mobile traffic is much lower than it was on desktop and so you go gosh, that's. Potentially not a very favorable Trend and we we for sure talk with that with the are friends from Adobe around the holiday episode but I think I think it's coming up on a couple shows and I was actually surprised to find out that we haven't done a deep dive so maybe that something will. Will do it at future show but I know you and I have done a number of live presentations were weave weave debated the Mobil gas. [25:44] And I guess what I would say to Arya a couple of things. [25:52] Most clients if you looked you mentally at your desktop and your Mobile Traffic your traffic probably supposed to go inside look up there traffic isn't flat there traffic is actually increasing. It's a one of the things that's happening is some of those mobile visits that don't convert well are incremental visits. And of course there's a because it's so much harder to buy something on a mobile device there's a lot more friction to check out. There's way you know West support for plugins in your browser so your payment information is less likely to be. Be stored in there and we joke a lot about a taking three hands to check out on a mobile device right one told the phone one that tap the virtual keyboard and 1/3 to hold your credit card. [26:36] The that that friction you know makes it less likely that people check out people also on mobile devices are are generally in a more micro moment context. They might be at the red light in the light turns green or they might be in line at the bank and get to the front of the bank or you know they might be doing something. Weather going to get interrupted in the much shorter. Of time so all that friction leads to too much more abandonment and so. We are seeing things where we're at experiences that reduce the friction improve the mobile Gap they don't make it go away but they you know if you look at the best mobile checkouts they have our mobile free mobile apps then. Then the traditional bad mobile checkouts have also a percentage of that is. [27:25] Not real inserted incorrectly measuring conversion so so most sides you know. Back to the simple formula conversion how many people bought versus how many people visited the site and of course mobile gives people a bunch of new reasons to visit your site so a bunch of mobile customers are coming to find out your store hours or if you have something in stock. Or what store is near than those are all things that used to do with the Yellow Pages in the in the analog phone. And with the newspaper in those visits are not coming to your site that customer had no intention to buy online they're ultimately going to go to your store. It does look like non-converting Mobile customer so so some of its an attribution problem and then the last thing we talked about is this Multi-Device attribution problem where. Because it is harder to check out on a mobile phone a lot of people will build their list do their pulmonary shopping on mobile. And then they'll ultimately consummate the purchased on their they're desktop browser where they you know are more likely to have payment information stored or or use a keyboard. [28:28] A password Plug-In or something like that that that makes it easier to pay. And because of the way that because we use cookies when you come back on your desktop you don't mess in your not authenticated as much users are. You look like a different visitor than the visitor that came on mobile so instead of it looking like got window came to my site twice and bought on the second visit it looks like. Got Wingo number one came to my side and mobile and didn't buy and some unrelated Scot Wingo came to my site later on a desktop and did by the. Yeah I don't think that's the the dominant mode but that absolutely is a mode and interesting Lee it at Publix this week we built this database with that now has over 2 billion device IDs in it. That we can map back to individual users and sure enough view you see if there still is a pretty substantial. A chunk of Christ of a shopping happening on a bunch of these e-commerce site so. All of that is interesting but here's the real bad news. You asked the great question at the end if that's the trend then how is e-commerce growing e-commerce should be shrinking everyone's moving a mobile and mobile doesn't come out as well why is he Commerce not drinking. And the bad reason for that is because not every site suffers from the mobile Gap. And the sites that don't suffer from the mobile Gap are you and the biggest most dominant sites in the markets right so. [29:59] Well no sides have a very low percentage of authenticated users Amazon has a very high percentage of authenticated users, and by all accounts has a very healthy mobile conversion rate right and so you have some of those sites at the top of the echo system that have a disproportionate Cent percent of the traffic and sales also way outperform the industry averages in Mobile and that is driving a lot of the e-commerce growth. Scot: [30:27] Yeah yeah we could probably do a whole show on this so I'll just kick it the next question before I get into a controversial topic that we have to go back. Jason: [30:36] So are you saying that was not a good lightning round answer. Scot: [30:39] That was very good and I'm not going to ruin it by it by adding on third question is from Alexandra volakis. I said it's about omni-channel so this is another one that squarely in your Wheelhouse in a centrally. And I'm kind of tripping this little bit how do you decide where it's best to ship from so I think what what L Alexander Andrew is kind of thinking about is you get an online order you've got ship from store or you've got a moment Center. [31:10] You probably have some complexity there you probably have you know Boosie's on each other guys have hundreds if not thousands of stores that could ship the product and then you have like let's say you have 5 phone at centers. [31:21] What's the what's the logic you would kind of work with a retailer to think about that do you just kind of go. Product is closer to the consumer here or ship from there or do you kind of is it cheaper to ship from the store or is it more expensive and how should people think about that. Jason: [31:37] Yep that's a great question and most retailers that have gotten successful with pretty complicated, fulfillment channels where they have a lot of different choices, either because they're feeling from store have a lot of different fulfillment centers there they're all using pretty sophisticated software sometimes that even uses machine learning to build a model for deciding how to do for filma and so normally we we call those the solutions of order Management Systems, the big Enterprise ones all have like very robust logic in them, but at the end of the day that the way you're implementing that logic for most cases is you're actually thinking about three big factors you're thinking about the cost of a fill so you want to optimize the lowest cost to fulfill, you're optimizing for the customer experience in the customer experiences is generally two big factors one is how fast you can get it to that customer so you want to get to him as quickly as possible obviously and another is, you want to get multi-item orders to the customer together so you'd rather ship everything in one box, not only is that more economical many cases but it's also just a better customer experience then split shipping from multiple fulfillment centers and the third is this this notion of inventory potential. And that that can get a little more complicated but essentially what it amounts to is. [33:12] Whatever fulfillment vehicle you fulfill for this order is going to leave inventory in the other for film of vehicles and what is the likelihood of there being further demand for, that next piece of inventory so when you're getting really sophisticated you you may. Choose a film that vehicle that isn't your cheapest because, it's likely to be the only demand in that particular fulfillment set of vehicle and there's likely to be other demand in the other fulfillment channels that's even lower cost for the rest of your goods, so I'm not sure I explain that super clearly, but like at at one level or another you basically are are putting together an analogue Rhythm that that optimizes for that customer experience that potential the cell and that that cost of fulfillment and, you know there are both a number of Enterprise off-the-shelf tools that do that in there a lot of the custom software that a lot of retards have, built over time to do it. Scot: [34:16] I will kind of dispute one thing so I actually like it when when Amazon since we split orders and they send them to me that when the stress available at I think that's a better customer experience I don't think it's a better. It's cheaper, for the retailer but you kind of implied it's better customer experience get all your stuff together that assumes that all would come together but I think most times you're having to choose you know do your hold up so it's one of the least common denominator problem. Jason: [34:42] So great potential Nuance like I would certainly agree, that to a certain extent like if if I'm an option to get two things faster than the other things and I and then option is overtly presented to me and I choose to get them as fast as possible I agree with you I'm a shopper that appreciates that and so, best customer experience for each customer is probably defined differently one problem with that experience is it can get very complicated right, and until I always use the Amazon versus Jet analogy and Amazon tends to make all those decisions for you but they tell you what they are and Jet you know is, is it sort of in the middle of giving you the choice of all those decisions and letting you choose for yourself as it split shipping actually both companies kind of let you choose for yourself but, what that the more friction that's in that choice like you actually see conversion go down, but the bigger issue is you and I are the least typically e-commerce Shoppers in the world and so for the overwhelming majority of people that buy stuff online they don't understand any of the nuances of fulfillment they don't understand that there are multiple fulfillment center that have some of these goods and so for most users they simply believe that when they order three things that they're using together in a project, that that those three things are all coming from the same source and so when the the the seller chooses to split ship or even just drop ship from one of the items from a manufacturer and it arrives on a different day what we see is. [36:17] A huge influx in customer service calls so customer service calls on switch shipments are way higher because customers just think. Something got left off the order they ordered, shoes and running shorts and a running shirt and they're using all three to go for a run and only to arrive you must have forgotten to ship me the 3rd and they don't understand that the third is coming direct from the manufacturer or from a different Warehouse or from the store and so, you know for those customers it's a bad customer experience to split ship but for sure, I'll totally agree with you and there's an elegant way to offer that to the customer make them understand then the best customer experience reach customers whatever they choose. Scot: [36:57] Yeah and then another thing all Throne of this is I think the omni-channel dirty secret is this ship from store and buy online pickup in-store, usually kind of sucks because I don't think stores know what's in the store like past half the time so so you know. Show me my worst online shopping experiences have been shipped from store and buy online pickup in-store and you know the ship from store stuff goes wrong because they're stock-outs where they thought they had the widget that happens you know. Lot more than a fulfillment center they also have you know they always say well just walking around the store in someone's cart we don't know where it is but I think their inventory is just really really bad at in stores and then the other thing is wrong kind of having. [37:45] Stuff cuz you got this salesperson there and they're trying to you know. Imagine you're in the shoe department at one of these retailers and you have to know about the shoes and then some on my order comes in and there's got to be part of your day where now you're. [38:00] Pick Pack ship person so we get a fair I would say the things we actually get that are in stock. You know a pretty material 5 10% there's usually some kind of error like we've been sent someone else's stuff or they did leave something out or you know if that kind of thing so. I know there's this kind of glassy omni-channel all your problems are solved but I found that most people really just don't do this for a while what are there any industry stats that you see her on that or. Jason: [38:31] Oh yeah so you're for sure right then did most people when they first do it totally suck at it in the one thing I would say is that there is a maturity curve there and when people get over that curve and get good at it, the customer satisfaction with the experiences very high so I would say like the. The benefit of being excellent at both of those experiences at at ship from store or buy online pickup in-store. But the potential upside is is true and very high, it's. It's easy to do it poorly and most people start out doing poorly so first actor. You're you nail that in-store inventory is a huge problem industry-wide and retailers never. Like the primary impetus to have super accurate inventory was. Was really your balance sheet for the most part like people don't even like purchase based on their inventory levels in in many retail stores in the old days. Inside like these experiences are the first ones to really put pressure on inventory accuracy in the store. [39:37] Inventory accuracy is getting way better there's both both machine learning and newer inventory systems have made it much. Easier for stores to get better at store inventory most of the big retailers now both Target and Walmart have robots running around the store taking pictures of shelves. And they're taking inventory based on this picture so they've actually taking people out of the equation, we're starting to see some new store concept that have intelligent shells so they can actually the shells take their own inventory and no right when there's out of stocks and things like that so the future of inventory accuracy is getting better, but at the end of the day almost every retailer I've ever worked with it started as a ship from store program started out with horrific metrics and so you know usually you have this, this error code item not found and you have this you don't sort of, percentage fulfillment like of all the orders I sent to a store what what percentage got filled and it's, totally common to see 50% of in-store orders be item not found or you know only be able to have a 50% fill rate when you first start shipping from store for, because of the inventory issues in the Employee Staffing and incompetency issues and I'll and the customers having the inventory and their Card issues all those things you can have huge failure rates in there and that creates a hideous customer experience, I've seen 90% item not found or 10% fill rates in some customers when they first want ship from store. [41:09] But if you many of the same customers I worked with it started out at 50% fill rates are now it like 94 96% fill rates, so over time they're able to put systems in places and process in place and be smarter about when they send the order to the store and trying not to fulfill when they have really thin inventory and only one in stock, and by implementing all those things the fill rate goes way up and you can today absolutely look at a Target and Best Buy and see that they're generating a meaningful economic advantage against Amazon, by being able to ship at a significant portion of their e-commerce but business from the store one zone get it to customers fast and cheap. Scot: [41:54] I'm learning a lot from this we should get that list of questions for Julia. Guitar chorus is the P silent or talk how do you think about singles day are there some retailers who take part of the vent in the US Amazon focuses more on Cyber week, and doesn't really do anything on single stay why is that. Jason: [42:20] So great question will be talking more about this I am not bullish on Singles day becoming a global holiday that's. Heavily the big factor here in the US and then there's a variety of reasons for that it already is a holiday in the u.s. is Veterans Day which is, somewhat problematic for turning it into a high-volume shopping day Alibaba just doesn't have a significant presence here at the moment, you know said they were they were years when when, Alibaba was having huge success in China and they're making noise about next year is going to be a much bigger Western holiday and. What that's morphed into in my mind and perhaps we'll have them on the show here in the near future to defend themselves is, they're they're making it a much bigger deal for us Brands largely to sell to, Eastern consume customers that are celebrating singles day so I think it's because singles to become a huge event for a lot of my clients for example, but it's because they're selling to customers in other markets it's not because they're selling in the US, all that being said you know I think it is possible to to create a new holiday here certainly Prime day is it is, a great example in the in the west but you know another interesting one is, Cyber Monday has become a very big holiday in Europe and as most of our listeners are probably aware. [43:57] They're not celebrating Thanksgiving in Europe so it is possible to create these shopping holidays I just think the dynamic of trying to create a holiday on Veterans Day a couple weeks before a very traditional shopping. you know for non-income and Company is is a rolling a rock up a pretty big hill. Scot: [44:21] And I will be a little facetious and Amazon does participate in. I'm single stamp but they do it in China so Amazon runs T-Mobile store in China and they saw other devices there and it just shows. China's interesting to me we talked about Amazon all on the show cuz it's the one area where Amazon has not been dominant then you argue with either the number three or four player in China. And it's because Alibaba has really kind of dominated with a. Different local way of doing things that that Amazon I wasn't able to replicate so because of that you have some really weird things that must be kind of painful for Amazon tap to do but example as they do sell Auntie mall now and then, they do accept Ali pay so this is the one. Region where you know Amazon doesn't control the entire payment world so I can the US they don't take Paypal because they have the power to kind of say no we want all that to flow through our system so. Little car fun fact for you if you didn't know that. Jason: [45:23] I did not know that the only part that's pretty funny. Scot: [45:28] And then the last question so Melissa Burdick another kind of friend of the show. How is the bankruptcy of Toys R Us going to impact Amazon this holiday is it going to be a bloodbath and pricing with, Toys R Us cutting prices at the stores because the bankruptcy in an Amazon matching and then this kind of Race To The Bottom. Jason: [45:49] So interesting question unless confident in my answer here but I think there's two, two issues is this holiday. Going to be a bloodbath of discount pricing number like regardless of Toys R Us like are a bunch of retards going to start you know early and aggressive discounts and is that going to drive. Pricing down for the whole holiday. I think it's an open question and frankly I'm very nervous about that like all of the the early forecast for Holiday are for, for pretty significant growth and robust sales and the unspoken truth in a lot of those is most years we have that kind of growth it's because, we sold stuff really cheap and discounted really deeply and potentially because we had too high of an inventory position and then you don't have to Discount more deeply, so I think the fact that there been a bunch of bankruptcies and More Store closures than usual this year and more distressed inventory his, has flooded the market and that that's cause more inventory full price inventory to get abandon on the shelves so I do think we're going to go into this holiday season with retailers in a little bigger inventory position than they'd like. And so I'm just frankly concerned overall, that then it's going to be heavily promotional holiday. What we already seen some early indication that it was going to start their sales super early, so all of those things could just turn it into a bloodbath not because of Toys R Us current current bankruptcy status. [47:19] Actually think. The Toys R Us in the current status has a disincentive to aggressively promote like the stores have to operate profitably over a holiday, and so I think they're not going to be the first one to drop their drawers on price like I think if they become really aggressive and promotions it's going to be later in the season, as they see how the the holiday is is shaking up but I think the. At this point they're not looking to liquidate inventory or those kinds of things like I think that's, you know if they decide they have to close 300 stores and they hire Gordon Brothers 2 to come in and liquidate inventory like that that potentially create a bloodbath but I don't think that's going to happen until 2018, so I kind of suspect toys is not going to be the the fuse that lights the. The the discounting fuse but I'm not sure that we we aren't going to see a bloodbath nonetheless. Scot: [48:24] Yeah. I would just add I'm an e-commerce software guy and I've learned a lot about retail over the years that I didn't know and I know Melissa used to work at Amazon so she. Definitely got kind of a similar kind of DNA on the digital side. And does really good Bloomberg article that will link to in the show notes that talks about how all these retailers these traditional retailers have really loaded up on debt, and you know what what happens is they get acquired by private Equity Firm and part of their model is to take the Assets in leverage them pretty highly meaning piling on a fair amount of debt. And what does this done is left. The entire segment pretty exposed to a Destructor like Amazon because in in Toys R Us has a good case study that you brought up, so Toys R Us has something like four or five billion dollars in debt and this debt comes in these tranches so you have all that dead out there and I'll have maturity dates and, Toys R Us couldn't actually deal with about 400 million of that which is what pushed him into bankruptcy so what happens is when you yo so Amazon has no debt and you a lot of retailers argue that she doesn't even care they don't make a profit. Talk about down the show but what what happens if you have a competitor like that. Come in and make a pretty small impact on you so maybe you'd lose five or 10% of sales doesn't feel like that would really. Australia upside down. What's Insidious is Amazon it knows everyone's margin because they have all this data and you may lose 5 or 10% of sales but that's probably your most profitable stuff and maybe lose 15 or 20% of a profit. [50:02] Are Eva. And that's what this debt is all priced against is. [50:07] 1015 years ago when this debt was piled on everyone assumed that your profit margin would be the same. [50:14] And then you have a new competitor, long. And they're able to Chisel a enough profit that it really tips you over so this article doesn't really good job of kind of, because really in-depth and looks at at that which is pretty interesting and has a whole map that shows kind of the hot areas and the whole point of the article. Apocalypse is just getting started to date when will you get from a debt perspective it looks like we're just at the beginning of a bloodbath the thing I've learned in this was too I guess we had on the show is. These at the mall level Aldi's anchor tenants effectively don't pay much in rent and there are because the word anchor they're there to draw other people in, what happens if so let's say you're a small mob a store and one of the anchors goes out of business usually is written in the lease that. Because you were drawn there by an anchor if an anchor leaves you. You are now free from your lease so these balls are unwinding at a pretty incredible pace and. There is a I don't follow it that closely but there's a lot of rumors that some of them are going to be sold and the mall for large Mall reach because they are in such a stress situation so. So so this kind of gets Amplified didn't did these things are not mutually exclusive so now you have stores at malls that are anchors and have huge debt and if it's caused this kind of death spiral that's happening there. At the mall level is kind of what I called Mulligan so interesting things that I wouldn't have learned about until the podcasting and try to understand what is going on out there. Jason: [51:47] That's why everyone should start a podcast. Scot: [51:49] Absolutely or listeners last question so this is from James lecourt how do you see augmented reality playing a role in e-commerce and when do you think it will be mainstream and accessible to the smaller retailers. Jason: [52:04] Another interesting question James we've done aviare are deep dive and I think Scott and I are sort of an alignment like VR is, truly interesting for some other reasons but I actually don't think it's it in the near-term very relevant to e-commerce I think augmented reality is potentially. Way more relevant but it's I think most of the use cases in Commerce are actually digital in-store use cases. [52:33] So warning more about getting more of the digital content to learn about products when you're in a physical store augmented reality in e-commerce the big use cases you think about are things like. How will that art look in my house will that furniture fit in my house what. You know what would this clothes look like on a a virtual representation of me or me and this me or these kinds of things. And what's interesting the. Rudimentary version of that technology is all out there oh I should mention the like virtual makeup stuff when beauty stuff which is has become quite good. So that the technology is all out there it's involving very quickly and so both Google and apple have really robust. New AR kits building in the latest version of their operating systems. And you look at the kind of experiences you can have on those those devices these latest devices that are using these they are kids. And you go man that's really compelling so if you have an iPhone 8 or an iPhone 10 I'd highly recommend you download this app called House craft. And how scrap uses AR to place Furniture in your house and it's it's amazing it's much better than some of the rudimentary stuff you've seen from. Some of the retailers Warby Parker has already leverage the AR kit in their app. [54:07] For virtual try-on of sunglasses and so you think about the face recognition technology that's in the iPhone 10 and the hundreds of measurements it's taking your face. Warby Parker take out of the AR kid all of those measurements put them into a deep Learning System, in recommend sunglass frames to you that are best suited to your face, and it creates an amazing AR experience and so you look at those things and you go man that is the future that that really is going to become mainstream, but then there's a big Debbie Downer in terms of how fast it's all going to happen, does AR kids only work on a small percentage of the hardware that real people own right so it it only runs on the the latest and greatest Hardware, so we have to wait for a couple upgrade Cycles to everyone, I want to get up to that that hardware and then at the moment those best experiences are really only deliverable through apps, and we've talked about this a lot on the show as well but for most retailers in for sure for small shops, it's next to impossible to get a a meaningful volume of customers to download and use your Mobile app and so what we really need is this robust AR capability to to be available in the web browser, not in the app and it is coming it's just still probably a couple years away so I think right now we're at the point where. [55:40] On the Best Hardware in an app customers are seeing experiences that really can drive conversion and sell more stuff and I think we're going to see more examples like the Warby Parker app that are going to be very very persuasive but it's probably another 3 years before, the majority of consumers have that capability in a web browser and that's when it becomes really meaningful for those medium and small size shops. Scot: [56:03] Yeah and I would add another challenge for a small size shop is the 3D models so, to put your products into this 3D World you have to have models of them and this is not a trivial skill set for folks to have and there's not a great solution for just kind of imagine you ran I don't know, sports store and you wanted to put everything into a virtual world there's no really good off-the-shelf solution for kind of scanning that stuff that, a mere mortal can handle and build the models so that's another one is like how do you partner with this assume your retailer multi-brand retailer you're going to partner with your Brands and they're going to have to have a level of sophistication where you call and say hey I really need 3D models for all your stuff they're going to have to know what you're talking about how many brand struggle just to get you the. Current tenant to the digital assets so that that's going to be an interesting challenge to see who solves that because you could end up in this scenario, because worst laws applying all these other things that pretty quickly we get the hardware as Jason mentioned all that stuff solved and it's pretty easy for you to have a platform but you just don't have the assets. Jason: [57:11] Yep and although I would point out just got you take that the that iPhone 10 sensor array in the Notch and you put it on the turntable and you suddenly have a pretty darn good cheap 3D scanner in so, you know, you you could imagine that the ability to to 3D scan and very high-quality at very well cost is something that Moore's Law is also going to deliver to us over the next two or three years. Scot: [57:38] Cooper we really appreciate it when asking the questions there and we have about 5 or 10 minutes to catch up on news and it wouldn't be a Jason Scott show without Amazon news. [58:03] Jason you had let's kick it off of you you walked into your Whole Foods was that today or yesterday and you had an interesting situation tell us about it. Jason: [58:15] Yeah that was today so there's a nice, two-story Whole Foods in my neighborhood downstairs is a very fancy coffee shop upstairs at the store and when I walked in the store today a big chunk of the coffee shop has been taken up by these. Temporary walls with all this Amazon signage and it looked like they were implying some kind of shopping experience was coming and that I got a chance to talk to some of the, the employees that we're doing it and it turns out they are this is a permanent installation. It's going into a bunch of Amazon stores in Chicago and it it basically is a Amazon device store, it's going into Whole Foods so another you know men's staffed place where you can go and get a echo demo or a Kindle demo or a fire. Demo and it sounds like they're going to have inventory for sale in the store and in ready to go. Scot: [59:14] Cool Sorry Amazon bookstore has that like little apple like section so you're kind of fishing it'll be like that couple tables. Jason: [59:21] Yeah and in fact I got to see the pictures and they look like they're straight out of the Amazon store. Scot: [59:26] And then there's also a nursing news where I forget who broke this but, Amazon is doing the nursing thing so if your third-party seller and Amazon to text that your price is a competitive what they'll do is they'll actually discount it and it says, sold by the third party seller but then discount provided by Amazon so, you know Amazon is pretty well known in history that they monitor prices across the internet in near real-time so I think what's happening there is they probably realize they were expensive in a couple areas especially that, part of the curve where they rely on third parties to sell things someone exclusively and they decided they. Wanted to not be disadvantaged there so they're actually funding that and it's a nursing so you had on average are going to pay as a third-party seller you pay Amazon 10% but then they kind of are selectively say you're effectively In-N-Out. The ones I've seen have been under 10% but they couldn't hear you actually go beyond that and say that we want to be competitive enough here that will fund even pass what the third-party is selling to us. [1:00:37] Actually figure out how to do a Google Search and Google Nexus Amazon. Pretty much real time and I found about 3000 items that that had this set so this is out of Amazon's like four or five hundred million this is not a huge thing at this point but the thing I thought you would find interesting is. Everyone I can I did look at all three thousand but I page through pretty quick they were all in the beauty category. Jason: [1:01:01] Yeah it it was super interesting and clever so obviously is as most of the regulars that showed no Amazon has a pretty sophisticated pricing I'll grab them on their 1p product, and you know when they sense a competitive situation they're they're very likely to be a fast follower, and they they see a lot of advantage in overall customer lifetime value even if they have to sell something at very narrow margins or even negative margins in the short term and you know the liability traditionally of the marketplaces you, Amazon doesn't have control over the pricing of that 3p product and so then you think about hey what are some categories that Amazon doesn't compete in in 1 p, but would really like to control prices in the 3p and you know there's certain kinds of products that that are tougher Amazon, and one of them would be like private label Cosmetics that have no interest in selling on Amazon but they want a third parties and gray marketers will by and and list on the Amazon market so that could actually be like, Ulta products. That should be exclusive to Ulta that's on the Amazon Channel and this tool gives them an opportunity to gives Amazon an opportunity to get really price competitive on that and, you know and in many cases that grey market product. Like the sellers are relying on selling because of convenience and so they often aren't super price competitive so this is a way for for Amazon to offer a competitive price in those categories were. [1:02:32] It wants to compete in the long run so that that's pretty clever but was interesting is there's a bunch of. [1:02:39] Not obvious unintended consequences of this program and it it's going to be funny to watch them all play out so there's all these things you wouldn't think of that initially but I'm sure Amazon stock through. When is something like returns so you know the seller offered offered a cosmetic 450 bucks Amazon discounted at 2:40 bucks. The consumer only paid 40 when you return it. You know Amazon has to refund part of your money and that seller has to refund its at all that stuff has to work out. But another big one is some of those sellers. Either have the other authorized sellers of a product they very likely have you know are complying with some some pricing requirements from. From their supplier so they they might have agreed to offer prices. Only at map price minimum advertised price in Amazon potentially could be discounting below that minimum advertised price so even though the seller is complying with their their pricing agreements. They're involuntarily out of compliance with that agreement because of this Amazon discount and I think another scenario is. Sellers that a promise to offer the same price to multiple marketplaces and then Amazon discounts at so effectively they're no longer complying with that agreement and so. You know those are all going to be some some that potentially sticky situations it's going to be interesting to see if any manufacturers come go after their sellers. [1:04:16] As a result of Amazon's price Judo. Scot: [1:04:21] Yeah yes, interesting to see how this plays out their solutions to these things so if you know you can imagine that if you don't want Amazon to do this there could be a knocked-out kind of thing and they've done that with a bunch of other programs, all the things are solvable but it is pretty interesting to see Amazon do this and you're there must have been some pressure in the beauty category that cause them to think about doing this. Jason: [1:04:45] What other things maybe is an opportunity for someone out there like maybe Channel advisor should do it but if you were a seller that used to be selling that good at, 50 bucks and you were competing against guys that were selling it at 60 and you suddenly see that Amazon is dropping t

united states tv jesus christ ceo amazon founders thanksgiving chicago australia europe google israel china disney internet house men sports woman talk law star wars west deep building holiday gold solo brand market microsoft dna western expectations iphone current fashion 3d press target stone product wall street walmart cbs raise android behavior vr shifting weather mobile id commerce saga wanted apocalypse taiwan paypal brands ipads prime trend ebay easier pricing rhythm consumer peak jedi bloomberg enterprise obsessive compulsive disorder industrial guitar cyber throne marketplace region niche kindle cent sellers demo today show assets discounts gap fast and furious cycles adobe trilogy singles mall raleigh hardware listener questions whole foods last jedi mid outsourcing temporary veterans day goldberg salmon jet cyber monday furniture warehouses alibaba misses best buy ro usps t mobile canyon aldi judo executive chairman scot cosmetics ids toys r us bob iger sephora insidious nuance auntie heavily under armour google search mulligan boosie shoppers mobil notch eerie publix in n out management systems d2c single day warby parker amplified kinect mama bears yoy tru a3 ulta yellow pages sporting goods kimberly clark sku debbie downers ryan johnson wayfarers salvia wheelhouse singles day chisel stony under armor predominantly jason scott rivet destructor capitalization fashion world razorfish branford 3d world apple hardware eclat us amazon amazon private label amazon news microsoft kinect ebay amazon new ar old market rudimentary skylake nitch force it scot wingo uniden nike free

CultCast #298 - iPhone 8... keynote date!

Play Episode Listen Later Aug 25, 2017 66:34

This week: Even more juicy details leak on iPhone 8, plus a report says the keynote reveal is right around the corner. Leander shares the strange twist in Apple’s autonomous car project Why the 13” MacBook Pro might soon become your favorite Mac Story time with L Kahney - he’s going to share the highlights from his whirlwind tour of Japan and other defenseless Asian lands. Plus Erfon recalls what it’s like in the darkness of a total solar eclipse. This episode supported by CultCloth will keep your iPhone 7, Apple Watch, Mac and iPad sparkling clean, and for a limited time you can use code CULTCAST to score a free CleanCloth with any order at CultCloth.co. The Cult of Mac watch store has the best straps in the biz. Thanks to Kevin MacLeod at incompetech.com for the great music you hear on today's show. On the show this week @erfon / @bst3r / @lkahney This week’s intro (thanks Nick Bracken!) http://www.ngataonga.org.nz/collections/catalogue/catalogue-item?record_id=147476 The shadows take on the sun's new shape https://instagram.com/p/BYEHnqbHU1d/ We might know the iPhone 8 keynote date https://www.cultofmac.com/498836/iphone-8-release-date/ Hot on the heels of Samsung’s grand unveiling of the Galaxy Note 8 today, carrier sources have supposedly informed Mac4Ever that Apple plans to unleash its device in just a few weeks. Apple’s keynote is allegedly slated for September 12 where the company will reveal three new iPhones. While we haven’t been able to verify the accuracy of the report ourselves, Apple usually hosts its iPhone keynote in the middle of September so the date could make sense. The French Apple blog claims that carriers have been informed by Apple to expect the device announcement on the second Tuesday of next month. Carriers usually receive a heads up so they can start planning marketing and to organize pre-orders to ensure inventory. And get this, the minimum storage capacity for Apple's OLED iPhone is said to be 64GB, with a 256GB option offered as the mid-tier capacity and a 512GB option at the highest tier, while 3GB of RAM is claimed to be included across the board. iPhone 8’s amazing facial recognition is super quick, works in the dark https://www.cultofmac.com/498426/iphone-8s-amazing-facial-recognition-super-quick-works-dark/ The iPhone 8’s facial recognition feature will work in a millionth of a second, and be more secure than the existing Touch ID sensor, and even work in the dark, a pair of new reports claim. In addition to the regular iPhone sensors you’d expect to find, the upcoming handset will reportedly boast a new “structured light” sensor, which uses bounced infrared light to work out the depth of different points on the face. That information is then used to build a 3D mesh of objects, which is compared to the one recorded when setting up the new iPhone. Calculating the timing between when infrared light is sent out and recorded coming back will let the iPhone work out accurate depth measurements. This, in turn, means you won’t be able to trick the handset using a 2D photo. The facial recognition is reportedly powered by tech Apple acquired when it bought Kinect motion sensor maker PrimeSense several years ago. The speed that the iPhone 8 facial recognition sensor will reportedly work is particularly impressive. The new handset will allegedly be able to do all of this within “a few hundred milliseconds,” which would make it a faster means of unlocking your iPhone than the current-generation Touch ID, Apple has been using Touch ID since 2013’s iPhone 5s. Apple’s 3D sensing tech is two years ahead of the competition https://www.cultofmac.com/498524/apples-3d-sensing-technology-two-years-ahead-competition/ According to a new report from KGI Securities analyst Ming-Chi Kuo, Apple competitor Qualcomm is working on its own 3D sensing tech, but it’s at least two years behind. And handset-makers aren’t in a rush to embrace it quite yet. "While Qualcomm is the most engaged company in the R&D of 3D sensing for the Android camp, a number of issues plague Qualcomm that prevent its tech from being ready for mass-market products. Immature algorithms, and thermal problems" Apple autonomous car morphs into self-driving shuttle bus for employees https://www.cultofmac.com/498760/apples-autonomous-car-morphed-self-driving-shuttle-bus-employees/ According to the New York Times, Apple’s secretive “Project Titan” self-driving car project has switched gears, transforming into an effort to build a self-driving shuttle bus. Called Palo Alto Infinite Loop, or PAIL, the shuttle would carry Apple employees between buildings. The project may serve as a test bed for Apple’s autonomous car research. But a customer-focused vehicle built by Apple is for now reportedly out of the question. Instead, Apple’s self-driving technology will likely be used by other carmakers eventually. The newspaper claims a leadership clash hampered project. Steve Zadesky, an Apple executive initially in charge of Titan, wanted to build semiautonomous technology. Meanwhile, Apple design chief Jony Ive “believed that a fully driverless car would allow the company to reimagine the automobile experience.” Apple reportedly investigated several innovative ideas for the project. Those included motorized doors that opened and closed silently, augmented reality displays for the interior of the car, new ways of incorporating the light sensor essential to driverless cars, and a total lack of steering wheel and gas pedals. Apple also researched the possibility of using globelike wheels for the vehicle, “because spherical wheels could allow the car better lateral movement.” Intel Launches First Eighth-Generation Core Processors, Paving Way For Quad-Core 13-Inch MacBook Pro https://www.macrumors.com/2017/08/21/intel-announces-8th-gen-core-kaby-lake-refresh/ The first four eighth-generation processors launching today are U-series chips suitable for the 13-inch MacBook Pro, MacBook Air, and Mac mini. They're all 15W chips with four cores and eight threads, paving the way for a quad-core 13-inch MacBook Pro should Apple choose to release one. The eighth-generation Core i5 and Core i7 chips are up to 40 percent faster than the equivalent seventh-generation Kaby Lake processors First MacBook Pro with Touch Bar uses a 6th gen Skylake processor. Intel also boasted that its eighth-generation Core processors are up to twice as fast as its equivalent five-year-old Ivy Bridge chips. It said users can output a 106-second 4K video in as little as three minutes with a new PC, for example, versus up to 45 minutes on an equivalent five-year-old PC.

apple japan new york times iphone 3d asian pc android cult mac ipads intel ram samsung kevin macleod keynote apple watches 4k 2d r d macbook pro qualcomm calculating carriers macbook air immature kinect galaxy note jony ive leander touch id iphone8 touch bar pail project titan 64gb 256gb ming chi kuo 512gb 3gb inch macbook pro 15w skylake kaby lake ivy bridge cultcast primesense kgi securities

204: WWF - Wayland, Weston, and FreeBSD

sizes round one cray market conditions epyc skylake

Play Episode Listen Later Jul 26, 2017 81:10

In this episode, we clear up the myth about scrub of death, look at Wayland and Weston on FreeBSD, Intel QuickAssist is here, and we check out OpenSMTP on OpenBSD. This episode was brought to you by Headlines Matt Ahrens answers questions about the “Scrub of Death” In working on the breakdown of that ZFS article last week, Matt Ahrens contacted me and provided some answers he has given to questions in the past, allowing me to answer them using HIS exact words. “ZFS has an operation, called SCRUB, that is used to check all data in the pool and recover any data that is incorrect. However, if a bug which make errors on the pool persist (for example, a system with bad non-ecc RAM) then SCRUB can cause damage to a pool instead of recover it. I heard it called the “SCRUB of death” somewhere. Therefore, as far as I understand, using SCRUB without ECC memory is dangerous.” > I don't believe that is accurate. What is the proposed mechanism by which scrub can corrupt a lot of data, with non-ECC memory? > ZFS repairs bad data by writing known good data to the bad location on disk. The checksum of the data has to verify correctly for it to be considered "good". An undetected memory error could change the in-memory checksum or data, causing ZFS to incorrectly think that the data on disk doesn't match the checksum. In that case, ZFS would attempt to repair the data by first re-reading the same offset on disk, and then reading from any other available copies of the data (e.g. mirrors, ditto blocks, or RAIDZ reconstruction). If any of these attempts results in data that matches the checksum, then the data will be written on top of the (supposed) bad data. If the data was actually good, then overwriting it with the same good data doesn't hurt anything. > Let's look at what will happen with 3 types of errors with non-ECC memory: > 1. Rare, random errors (e.g. particle strikes - say, less than one error per GB per second). If ZFS finds data that matches the checksum, then we know that we have the correct data (at least at that point in time, with probability 1-1/2^256). If there are a lot of memory errors happening at a high rate, or if the in-memory checksum was corrupt, then ZFS won't be able to find a good copy of the data , so it won't do a repair write. It's possible that the correctly-checksummed data is later corrupted in memory, before the repair write. However, the window of vulnerability is very very small - on the order of milliseconds between when the checksum is verified, and when the write to disk completes. It is implausible that this tiny window of memory vulnerability would be hit repeatedly. > 2. Memory that pretty much never does the right thing. (e.g. huge rate of particle strikes, all memory always reads 0, etc). In this case, critical parts of kernel memory (e.g. instructions) will be immediately corrupted, causing the system to panic and not be able to boot again. > 3. One or a few memory locations have "stuck bits", which always read 0 (or always read 1). This is the scenario discussed in the message which (I believe) originally started the "Scrub of Death" myth: https://forums.freenas.org/index.php?threads/ecc-vs-non-ecc-ram-and-zfs.15449/ This assumes that we read in some data from disk to a memory location with a stuck bit, "correct" that same bad memory location by overwriting the memory with the correct data, and then we write the bad memory location to disk. However, ZFS doesn't do that. (It seems the author thinks that ZFS uses parity, which it only does when using RAID-Z. Even with RAID-Z, we also verify the checksum, and we don't overwrite the bad memory location.) > Here's what ZFS will actually do in this scenario: If ZFS reads data from disk into a memory location with a stuck bit, it will detect a checksum mismatch and try to find a good copy of the data to repair the "bad" disk. ZFS will allocate a new, different memory location to read a 2nd copy of the data, e.g. from the other side of a mirror (this happens near the end of dslscanscrub_cb()). If the new memory location also has a stuck bit, then its checksum will also fail, so we won't use it to repair the "bad" disk. If the checksum of the 2nd copy of the data is correct, then we will write it to the "bad" disk. This write is unnecessary, because the "bad" disk is not really bad, but it is overwriting the good data with the same good data. > I believe that this misunderstanding stems from the idea that ZFS fixes bad data by overwriting it in place with good data. In reality, ZFS overwrites the location on disk, using a different memory location for each read from disk. The "Scrub of Death" myth assumes that ZFS overwrites the location in memory, which it doesn't do. > In summary, there's no plausible scenario where ZFS would amplify a small number of memory errors, causing a "scrub of death". Additionally, compared to other filesystems, ZFS checksums provide some additional protection against bad memory. “Is it true that ZFS verifies the checksum of every block on every read from disk?” > Yes “And if that block is incorrect, that ZFS will repair it?” > Yes “If yes, is it possible set options or flag for change that behavior? For example, I would like for ZFS to verify checksums during any read, but not change anything and only report about issues if it appears. Is it possible?” > There isn't any built-in flag for doing that. It wouldn't be hard to add one though. If you just wanted to verify data, without attempting to correct it, you could read or scan the data with the pool was imported read-only “If using a mirror, when a file is read, is it fully read and verified from both sides of the mirror?” > No, for performance purposes, each block is read from only one side of the mirror (assuming there is no checksum error). “What is the difference between a scrub and copying every file to /dev/null?” > That won't check all copies of the file (e.g. it won't check both sides of the mirror). *** Wayland, and Weston, and FreeBSD - Oh My! (https://euroquis.nl/bobulate/?p=1617) KDE's CI system for FreeBSD (that is, what upstream runs to continuously test KDE git code on the FreeBSD platform) is missing some bits and failing some tests because of Wayland. Or rather, because FreeBSD now has Wayland, but not Qt5-Wayland, and no Weston either (the reference implementation of a Wayland compositor). Today I went hunting for the bits and pieces needed to make that happen. Fortunately, all the heavy lifting has already been done: there is a Weston port prepared and there was a Qt5-Wayland port well-hidden in the Area51 plasma5/ branch. I have taken the liberty of pulling them into the Area51 repository as branch qtwayland. That way we can nudge Weston forward, and/or push Qt5-Wayland in separately. Nicest from a testing perspective is probably doing both at the same time. I picked a random “Hello World” Wayland tutorial and also built a minimal Qt program (using QMessageBox::question, my favorite function to hate right now, because of its i18n characteristics). Then, setting XDGRUNTIMEDIR to /tmp/xdg, I could start Weston (as an X11 client), wayland-hello (as a Wayland client, displaying in Weston) and qt-hello (as either an X11 client, or as a Wayland client). So this gives users of Area51 (while shuffling branches, granted) a modern desktop and modern display capabilities. Oh my! It will take a few days for this to trickle up and/or down so that the CI can benefit and we can make sure that KWin's tests all work on FreeBSD, but it's another good step towards tight CI and another small step towards KDE Plasma 5 on the desktop on FreeBSD. pkgsrcCon 2017 report (https://blog.netbsd.org/tnf/entry/pkgsrccon_2017_report) This years pkgsrcCon returned to London once again. It was last held in London back in 2014. The 2014 con was the first pkgsrcCon I attended, I had been working on Darwin/PowerPC fixes for some months and presented on the progress I'd made with a 12" G4 PowerBook. I took away a G4 Mac Mini that day to help spare the PowerBook for use and dedicate a machine for build and testing. The offer of PowerPC hardware donations was repeated at this years con, thanks to jperkin@ who showed up with a backpack full of Mac Minis (more on that later). Since 2014 we have held cons in Berlin (2015) & Krakow (2016). In Krakow we had talks about a wide range of projects over 2 days, from Haiku Ports to Common Lisp to midipix (building native PE binaries for Windows) and back to the BSDs. I was very pleased to continue the theme of a diverse program this year. Aside from pkgsrc and NetBSD, we had talks about FreeBSD, OpenBSD, Slackware Linux, and Plan 9. Things began with a pub gathering on the Friday for the pre-con social, we hung out and chatted till almost midnight on a wide range of topics, such as supporting a system using NFS on MS-DOS, the origins of pdksh, corporate IT, culture and many other topics. On parting I was asked about the starting time on Saturday as there was some conflicting information. I learnt that the registration email had stated a later start than I had scheduled for & advertised on the website, by 30 minutes. Lesson learnt: register for your own event! Not a problem, I still needed to setup a webpage for the live video stream, I could do both when I got back. With some trimming here and there I had a new schedule, I posted that to the pkgsrcCon website and moved to trying to setup a basic web page which contained a snippet of javascript to play a live video stream from Scale Engine. 2+ hours later, it was pointed out that the XSS protection headers on pkgsrc.org breaks the functionality. Thanks to jmcneill@ for debugging and providing a working page. Saturday started off with Giovanni Bechis speaking about pledge in OpenBSD and adding support to various packages in their ports tree, alnsn@ then spoke about installing packages from a repo hosted on the Tor network. After a quick coffee break we were back to hear Charles Forsyth speak about how Plan 9 and Inferno dealt with portability, building software and the problem which are avoided by the environment there. This was followed by a very energetic rant by David Spencer from the Slackbuilds project on packaging 3rd party software. Slackbuilds is a packaging system for Slackware Linux, which was inspired by FreeBSD ports. For the first slot after lunch, agc@ gave a talk on the early history of pkgsrc followed by Thomas Merkel on using vagrant to test pkgsrc changes with ease, locally, using vagrant. khorben@ covered his work on adding security to pkgsrc and bsiegert@ covered the benefits of performing our bulk builds in the cloud and the challenges we currently face. My talk was about some topics and ideas which had inspired me or caught my attention, and how it could maybe apply to my work.The title of the talk was taken from the name of Andrew Weatherall's Saint Etienne remix, possibly referring to two different styles of track (dub & vocal) merged into one or something else. I meant it in terms of applicability of thoughts and ideas. After me, agc@ gave a second talk on the evolution of the Netflix Open Connect appliance which runs FreeBSD and Vsevolod Stakhov wrapped up the day with a talk about the technical implementation details of the successor to pkgtools in FreeBSD, called pkg, and how it could be of benefit for pkgsrc. For day 2 we gathered for a hack day at the London Hack Space. I had burn't some some CD of the most recent macppc builds of NetBSD 8.0BETA and -current to install and upgrade Mac Minis. I setup the donated G4 minis for everyone in a dual-boot configuration and moved on to taking apart my MacBook Air to inspect the wifi adapter as I wanted to replace it with something which works on FreeBSD. It was not clear from the ifixit teardown photos of cards size, it seemed like a normal mini-PCIe card but it turned out to be far smaller. Thomas had also had the same card in his and we are not alone. Thomas has started putting together a driver for the Broadcom card, the project is still in its early days and lacks support for encrypted networks but hopefully it will appear on review.freebsd.org in the future. weidi@ worked on fixing SunOS bugs in various packages and later in the night we setup a NetBSD/macppc bulk build environment together on his Mac Mini. Thomas setup an OpenGrock instance to index the source code of all the software available for packaging in pkgsrc. This helps make the evaluation of changes easier and the scope of impact a little quicker without having to run through a potentially lengthy bulk build with a change in mind to realise the impact. bsiegert@ cleared his ticket and email backlog for pkgsrc and alnsn@ got NetBSD/evbmips64-eb booting on his EdgeRouter Lite. On Monday we reconvened at the Hack Space again and worked some more. I started putting together the talks page with the details from Saturday and the the slides which I had received, in preparation for the videos which would come later in the week. By 3pm pkgsrcCon was over. I was pretty exhausted but really pleased to have had a few days of techie fun. Many thanks to The NetBSD Foundation for purchasing a camera to use for streaming the event and a speedy response all round by the board. The Open Source Specialist Group at BCS, The Chartered Institute for IT and the London Hack Space for hosting us. Scale Engine for providing streaming facility. weidi@ for hosting the recorded videos. Allan Jude for pointers, Jared McNeill for debugging, NYCBUG and Patrick McEvoy for tips on streaming, the attendees and speakers. This year we had speakers from USA, Italy, Germany and London E2. Looking forward to pkgsrcCon 2018! The videos and slides are available here (http://www.pkgsrc.org/pkgsrcCon/2017/talks.html) and the Internet Archive (http://archive.org/details/pkgsrcCon-2017). News Roundup QuickAssist Driver for FreeBSD is here and pfSense Support Coming (https://www.servethehome.com/quickassist-driver-freebsd-pfsupport-coming/) This week we have something that STH readers will be excited about. Before I started writing for STH, I was a reader and had been longing for QuickAssist support ever since STH's first Rangeley article over three and a half years ago. It was clear from the get-go that Rangeley was going to be the preeminent firewall appliance platform of its day. The scope of products that were impacted by the Intel Atom C2000 series bug showed us it was indeed. For my personal firewalls, I use pfSense on that Rangeley platform so I have been waiting to use QuickAssist with my hardware for almost an entire product generation. + New Hardware and QuickAssist Incoming to pfSense (Finally) pfSense (and a few other firewalls) are based on FreeBSD. FreeBSD tends to lag driver support behind mainstream Linux but it is popular for embedded security appliances. While STH is the only site to have done QuickAssist benchmarks for OpenSSL and IPSec VPNs pre-Skylake, we expect more platforms to use it now that the new Intel Xeon Scalable Processor Family is out. With the Xeon Scalable platforms, the “Lewisburg” PCH has QuickAssist options of up to 100Gbps, or 2.5x faster than the previous generation add-in cards we tested (40Gbps.) We now have more and better hardware for QAT, but we were still devoid of a viable FreeBSD QAT driver from Intel. That has changed. Our Intel Xeon Scalable Processor Family (Skylake-SP) Launch Coverage Central has been the focus of the STH team's attention this week. There was another important update from Intel that got buried, a publicly available Intel QuickAssist driver for FreeBSD. You can find the driver on 01.org here dated July 12, 2017. Drivers are great, but we still need support to be enabled in the OS and at the application layer. Patrick forwarded me this tweet from Jim Thompson (lead at Netgate the company behind pfSense): The Netgate team has been a key company pushing QuickAssist appliances in the market, usually based on Linux. To see that QAT is coming to FreeBSD and that they were working to integrate into “pfSense soon” is more than welcome. For STH readers, get ready. It appears to be actually and finally happening. QuickAssist on FreeBSD and pfSense OpenBSD on the Huawei MateBook X (https://jcs.org/2017/07/14/matebook) The Huawei MateBook X is a high-quality 13" ultra-thin laptop with a fanless Core i5 processor. It is obviously biting the design of the Apple 12" MacBook, but it does have some notable improvements such as a slightly larger screen, a more usable keyboard with adequate key travel, and 2 USB-C ports. It also uses more standard PC components than the MacBook, such as a PS/2-connected keyboard, removable m.2 WiFi card, etc., so its OpenBSD compatibility is quite good. In contrast to the Xiaomi Mi Air, the MateBook is actually sold (2) in the US and comes with a full warranty and much higher build quality (though at twice the price). It is offered in the US in a "space gray" color for the Core i5 model and a gold color for the Core i7. The fanless Core i5 processor feels snappy and doesn't get warm during normal usage on OpenBSD. Doing a make -j4 build at full CPU speed does cause the laptop to get warm, though the palmrest maintains a usable temperature. The chassis is all aluminum and has excellent rigidity in the keyboard area. The 13.0" 2160x1440 glossy IPS "Gorilla glass" screen has a very small bezel and its hinge is properly weighted to allow opening the lid with one hand. There is no wobble in the screen when open, even when jostling the desk that the laptop sits on. It has a reported brightness of 350 nits. I did not experience any of the UEFI boot variable problems that I did with the Xiaomi, and the MateBook booted quickly into OpenBSD after re-initializing the GPT table during installation. OpenSMTPD under OpenBSD with SSL/VirtualUsers/Dovecot (https://blog.cagedmonster.net/opensmtpd-under-openbsd-with-ssl-virtualusers-dovecot/) During the 2013 AsiaBSDCon, the team of OpenBSD presented its mail solution named OpenSMTPD. Developed by the OpenBSD team, we find the so much appreciated philosophy of its developers : security, simplicity / clarity and advanced features. Basic configuration : OpenSMTPD is installed by default, we can immediately start with a simple configuration. > We listen on our interfaces, we specify the path of our aliases file so we can manage redirections. > Mails will be delivered for the domain cagedmonster.net to mbox (the local users mailbox), same for the aliases. > Finally, we accept to relay local mails exclusively. > We can now enable smtpd at system startup and start the daemon. Advanced configuration including TLS : You can use SSL with : A self-signed certificate (which will not be trusted) or a certificate generated by a trusted authority. LetsEncrypt uses Certbot to generated your certificate. You can check this page for further informations. Let's focus on the first. Generation of the certificate : We fix the permissions : We edit the config file : > We have a mail server with SSL, it's time to configure our IMAP server, Dovecot, and manage the creation of virtual users. Dovecot setup, and creation of Virtual Users : We will use the package system of OpenBSD, so please check the configuration of your /etc/pkg.conf file. Enable the service at system startup : Setup the Virtual Users structure : Adding the passwd table for smtpd : Modification of the OpenSMTPD configuration : We declare the files used for our Virtual Accounts, we include SSL, and we configure mails delivery via the Dovecot lmtp socket. We'll create our user lina@cagedmonster.net and set its password. Configure SSL Configure dovecot.conf Configure mail.con Configure login.conf : Make sure that the value of openfiles-cur in /etc/login.conf is equal or superior of 1000 ! Starting Dovecot *** OpenSMTPD and Dovecot under OpenBSD with MySQL support and SPAMD (https://blog.cagedmonster.net/opensmtpd-and-dovecot-under-openbsd-with-mysql-support-and-spamd/) This article is the continuation of my previous tutorial OpenSMTPD under OpenBSD with SSL/VirtualUsers/Dovecot. We'll use the same configuration and add some features so we can : Use our domains, aliases, virtual users with a MySQL database (MariaDB under OpenBSD). Deploy SPAMD with OpenSMTPD for a strong antispam solution. + Setup of the MySQL support for OpenSMTPD & Dovecot + We create our SQL database named « smtpd » + We create our SQL user « opensmtpd » we give him the privileges on our SQL database and we set its password + We create the structure of our SQL database + We generate our password with Blowfish (remember it's OpenBSD !) for our users + We create our tables and we include our datas + We push everything to our database + Time to configure OpenSMTPD + We create our mysql.conf file and configure it + Configuration of Dovecot.conf + Configuration of auth-sql.conf.ext + Configuration of dovecot-sql.conf.ext + Restart our services OpenSMTPD & SPAMD : SPAMD is a service simulating a fake SMTP server and relying on strict compliance with RFC to determine whether the server delivering a mail is a spammer or not. + Configuration of SPAMD : + Enable SPAMD & SPAMLOGD at system startup : + Configuration of SPAMD flags + Configuration of PacketFilter + Configuration of SPAMD + Start SPAMD & SPAMLOGD Running a TOR relay on FreeBSD (https://networkingbsdblog.wordpress.com/2017/07/14/freebsd-tor-relay-using-priveledge-seperation/) There are 2 main steps to getting a TOR relay working on FreeBSD: Installing and configuring Tor Using an edge router to do port translation In my case I wanted TOR to run it's services on ports 80 and 443 but any port under 1024 requires root access in UNIX systems. +So I used port mapping on my router to map the ports. +Begin by installing TOR and ARM from: /usr/ports/security/tor/ /usr/ports/security/arm/ Arm is the Anonymizing Relay Monitor: https://www.torproject.org/projects/arm.html.en It provides useful monitoring graph and can be used to configure the torrc file. Next step edit the torrc file (see Blog article for the edit) It is handy to add the following lines to /etc/services so you can more easily modify your pf configuration. torproxy 9050/tcp #torsocks torOR 9090/tcp #torOR torDIR 9099/tcp #torDIR To allow TOR services my pf.conf has the following lines: # interfaces lan_if=”re0″ wifi_if=”wlan0″ interfaces=”{wlan0,re0}” tcp_services = “{ ssh torproxy torOR torDIR }” # options set block-policy drop set loginterface $lan_if # pass on lo set skip on lo scrub in on $lan_if all fragment reassemble # NAT nat on $lan_if from $wifi_if:network to !($lan_if) -> ($lan_if) block all antispoof for $interfaces #In NAT pass in log on $wifi_if inet pass out all keep state #ICMP pass out log inet proto icmp from any to any keep state pass in log quick inet proto icmp from any to any keep state #SSH pass in inet proto tcp to $lan_if port ssh pass in inet proto tcp to $wifi_if port ssh #TCP Services on Server pass in inet proto tcp to $interfaces port $tcp_services keep state The finally part is mapping the ports as follows: TOR directory port: LANIP:9099 —> WANIP:80 TOR router port: LANIP:9090 —-> WANIP:443 Now enable TOR: $ sudo echo “tor_enable=YES” >> /etc/rc.conf Start TOR: $ sudo service tor start *** Beastie Bits OpenBSD as a “Desktop” (Laptop) (http://unixseclab.com/index.php/2017/06/12/openbsd-as-a-desktop-laptop/) Sascha Wildner has updated ACPICA in DragonFly to Intel's version 20170629 (http://lists.dragonflybsd.org/pipermail/commits/2017-July/625997.html) Dport, Rust, and updates for DragonFlyBSD (https://www.dragonflydigest.com/2017/07/18/19991.html) OPNsense 17.7 RC1 released (https://opnsense.org/opnsense-17-7-rc1/) Unix's mysterious && and || (http://www.networkworld.com/article/3205148/linux/unix-s-mysterious-andand-and.html#tk.rss_unixasasecondlanguage) The Commute Deck : A Homebrew Unix terminal for tight places (http://boingboing.net/2017/06/16/cyberspace-is-everting.html) FreeBSD 11.1-RC3 now available (https://lists.freebsd.org/pipermail/freebsd-stable/2017-July/087407.html) Installing DragonFlyBSD with ORCA when you're totally blind (http://lists.dragonflybsd.org/pipermail/users/2017-July/313528.html) Who says FreeBSD can't look good (http://imgur.com/gallery/dc1pu) Pratik Vyas adds the ability to do paused VM migrations for VMM (http://undeadly.org/cgi?action=article&sid=20170716160129) Feedback/Questions Hrvoje - OpenBSD MP Networking (http://dpaste.com/0EXV173#wrap) Goran - debuggers (http://dpaste.com/1N853NG#wrap) Abhinav - man-k (http://dpaste.com/1JXQY5E#wrap) Liam - university setup (http://dpaste.com/01ERMEQ#wrap)

united states time death apple interview guide germany italy berlin blog plan generation built lesson memory os pc ps cd basic windows setup wifi rare pe developers intel ram arm drivers rust storage how to inferno open source restart tor gb tutorials developed gpt linux area51 macbook vm enable cpu xiaomi orca usb c sql dragonfly macbook air mails qt g4 internet archive ssl bcs kde krakow ecc scrub modification mac mini blowfish saint etienne broadcom chartered institute unix configuration mysql goran nicest wayland digital ocean powerbook rfc bsd abhinav jim thompson nfs freebsd openssl smtp andrew weatherall uefi sth xss pcie new hardware imap mariadb zfs powerpc david spencer let's encrypt openbsd rc1 kde plasma pfsense x11 qat matebook rc3 rangeley skylake kwin netbsd vmm common lisp patrick mcevoy allan jude netgate bsds ixsystems certbot trueos tarsnap sunos matt ahrens dragonflybsd asiabsdcon

Episode 186: EPYC vs Skylake Round One; Cray Right-Sizes to Market Conditions

This Week in HPC

Play Episode Listen Later Jul 24, 2017 17:30

Episode 186: EPYC vs Skylake Round One; Cray Right-Sizes to Market Conditions by Intersect360 Research

187: Zombie Modifier (hak)

Rebuild

Play Episode Listen Later Jul 24, 2017 121:43

Hakuro Matsuda さんをゲストに迎えて、Macbook Pro, USB-C, Kaby Lake, Ryzen, 転職、SF映画などについて話しました。 Show Notes GopherCon 2017 Afuri Ramen - Portland Here are some more ways to bring MagSafe charging back to your USB-C MacBook Pro Apple may create a MagSafe to USB-C adapter Anker PowerPort+ 5 USB-C Power Delivery Benson Leung's review of Juiced Systems USB-C Karabiner-Elements: The next generation Karabiner for macOS Sierra The New Razer Blade Gaming Laptop Skylake, Kaby Lake chips have a crash bug with hyperthreading enabled Ryzenにまつわる2つの問題 Pentium FDIV bug AMD vs Intel Market Share The new iPad Pro A10X chip is the first 10nm TSMC chip Samsung and Google are finally on the same side as Apple in a new battle Apple Machine Learning Journal ゼロから作るDeep Learning Swift Creator Chris Lattner Leaves Tesla After Only Six Months in the Job Chris Lattner's Resume Fallout ファミコンに続いて、スーパーファミコンが小さくなって再登場！アクトレイザー Zygote District 9 The Girl with All the Gifts The Last Of Us - Naughty Dog Black Mirror | Netflix Apple previews new emoji coming later this year Dragon Emoji BLADE RUNNER 2049 – Trailer 2 くりぃむ有田「有田と週刊プロレスと」シーズン2

google apple girl gifts zombies fallout trailer bc e3 samsung blade runner sf resume amd macbook pro deep learning usb c tsmc magsafe ryzen modifier macos sierra chris lattner zygote skylake black mirror netflix gophercon kaby lake karabiner

202: Brokering Bind

arm new systems isc skylake

Play Episode Listen Later Jul 12, 2017 76:58

We look at an OpenBSD setup on a new laptop, revel in BSDCan trip reports, and visit daemons and friendly ninjas. This episode was brought to you by Headlines OpenBSD and the modern laptop (http://bsdly.blogspot.de/2017/07/openbsd-and-modern-laptop.html) Peter Hansteen has a new blog post about OpenBSD (http://www.openbsd.org/) on laptops: Did you think that OpenBSD is suitable only for firewalls and high-security servers? Think again. Here are my steps to transform a modern mid to high range laptop into a useful Unix workstation with OpenBSD. One thing that never ceases to amaze me is that whenever I'm out and about with my primary laptop at conferences and elsewhere geeks gather, a significant subset of the people I meet have a hard time believing that my laptop runs OpenBSD, and that it's the only system installed. and then it takes a bit of demonstrating that yes, the graphics runs with the best available resolution the hardware can offer, the wireless network is functional, suspend and resume does work, and so forth. And of course, yes, I do use that system when writing books and articles too. Apparently heavy users of other free operating systems do not always run them on their primary workstations. Peter goes on to describe the laptops he's had over the years (all running OpenBSD) and after BSDCan 2017, he needed a new one due to cracks in the display. So the time came to shop around for a replacement. After a bit of shopping around I came back to Multicom, a small computers and parts supplier outfit in rural Åmli in southern Norway, the same place I had sourced the previous one. One of the things that attracted me to that particular shop and their own-branded offerings is that they will let you buy those computers with no operating system installed. That is of course what you want to do when you source your operating system separately, as we OpenBSD users tend to do. The last time around I had gone for a "Thin and lightweight" 14 inch model (Thickness 20mm, weight 2.0kg) with 16GB RAM, 240GB SSD for system disk and 1TB HD for /home (since swapped out for a same-size SSD, as the dmesg will show). Three years later, the rough equivalent with some added oomph for me to stay comfortable for some years to come ended me with a 13.3 inch model, 18mm and advertised as 1.3kg (but actually weighing in at 1.5kg, possibly due to extra components), 32GB RAM, 512GB SSD and 2TB harddisk. For now the specification can be viewed online here (https://www.multicom.no/systemconfigurator.aspx?q=st:10637291;c:100559;fl:0#4091-10500502-1;4086-10637290-1;4087-8562157-2;4088-9101982-1;4089-9101991-1) (the site language is Norwegian, but product names and units of measure are not in fact different). The OpenBSD installer is a wonder of straightforward, no-nonsense simplicity that simply gets the job done. Even so, if you are not yet familiar with OpenBSD, it is worth spending some time reading the OpenBSD FAQ's installation guidelines and the INSTALL.platform file (in our case, INSTALL.amd64) to familiarize yourself with the procedure. If you're following this article to the letter and will be installing a snapshot, it is worth reading the notes on following -current too. The main hurdle back when I was installing the 2014-vintage 14" model was getting the system to consider the SSD which showed up as sd1 the automatic choice for booting (I solved that by removing the MBR, setting the size of the MBR on the hard drive that showed up as sd0 to 0 and enlarging the OpenBSD part to fill the entire drive). + He goes on to explain the choices he made in the installer and settings made after the reboot to set up his work environment. Peter closes with: If you have any questions on running OpenBSD as a primary working environment, I'm generally happy to answer but in almost all cases I would prefer that you use the mailing lists such as misc@openbsd.org or the OpenBSD Facebook (https://www.facebook.com/groups/2210554563/) group so the question and hopefully useful answers become available to the general public. Browsing the slides for my recent OpenBSD and you (https://home.nuug.no/~peter/openbsd_and_you/) user group talk might be beneficial if you're not yet familiar with the system. And of course, comments on this article are welcome. BSDCan 2017 Trip Report: Roller Angel (https://www.freebsdfoundation.org/blog/2017-bsdcan-trip-report-roller-angel/) We could put this into next week's show, because we have another trip report already that's quite long. After dropping off my luggage, I headed straight over to the Goat BoF which took place at The Royal Oak. There were already a number of people there engaged in conversation with food and drink. I sat down at a table and was delighted that the people sitting with me were also into the BSD's and were happy to talk about it the whole time. I felt right at home from the start as people were very nice to me, and were interested in what I was working on. I honestly didn't know that I would fit in so well. I had a preconceived notion that people may be a bit hard to approach as they are famous and so technically advanced. At first, people seemed to only be working in smaller circles. Once you get more familiar with the faces, you realize that these circles don't always contain the same people and that they are just people talking about specific topics. I found that it was easy to participate in the conversation and also found out that people are happy to get your feedback on the subject as well. I was actually surprised how easily I got along with everyone and how included I felt in the activities. I volunteered to help wherever possible and got to work on the video crew that recorded the audio and slides of the talks. The people at BSDCan are incredibly easy to talk to, are actually interested in what you're doing with BSD, and what they can do to help. It's nice to feel welcome in the community. It's like going home. Dan mentioned in his welcome on the first day of BSDCan that the conference is like home for many in the community. The trip report is very detailed and chronicles the two days of the developer summit, and the two days of the conference There was some discussion about a new code of conduct by Benno Rice who mentioned that people are welcome to join a body of people that is forming that helps work out issues related to code of conduct and forwards their recommendations on to core. Next, Allan introduced the idea of creating a process for formally discussing big project changes or similar discussions that is going to be known as FCP or FreeBSD Community Proposal. In Python we have the Python Enhancement Proposal or PEP which is very similar to the idea of FCP. I thought this idea is a great step for FreeBSD to be implementing as it has been a great thing for Python to have. There was some discussion about taking non-code contributions from people and how to recognize those people in the project. There was a suggestion to have a FreeBSD Member status created that can be given to people whose non-code contributions are valuable to the project. This idea seemed to be on a lot of people's minds as something that should be in place soon. The junior jobs on the FreeBSD Wiki were also brought up as a great place to look for ideas on how to get involved in contributing to FreeBSD. Roller wasted no time, and started contributing to EdgeBSD at the conference. On the first day of BSDCan I arrived at the conference early to coordinate with the team that records the talks. We selected the rooms that each of us would be in to do the recording and set up a group chat via WhatsApp for coordination. Thanks to Roller, Patrick McAvoy, Calvin Hendryx-Parker, and all of the others who volunteered their time to run the video and streaming production at BSDCan, as well as all others who volunteered, even if it was just to carry a box. BSDCan couldn't happen without the army of volunteers. After the doc lounge, I visited the Hacker Lounge. There were already several tables full of people talking and working on various projects. In fact, there was a larger group of people who were collaborating on the new libtrue library that seemed to be having a great time. I did a little socializing and then got on my laptop and did some more work on the documentation using my new skills. I really enjoyed having a hacker lounge to go to at night. I want to give a big thank you to the FreeBSD Foundation for approving my travel grant. It was a great experience to meet the community and participate in discussions. I'm very grateful that I was able to attend my first BSDCan. After visiting the doc lounge a few times, I managed to get comfortable using the tools required to edit the documentation. By the end of the conference, I had submitted two documentation patches to the FreeBSD Bugzilla with several patches still in progress. Prior to the conference I expected that I would be spending a lot of time working on my Onion Omega and Edge Router Lite projects that I had with me, but I actually found that there was always something fun going on that I would rather do or work on. I can always work on those projects at home anyway. I had a good time working with the FreeBSD community and will continue working with them by editing the documentation and working with Bugzilla. One of the things I enjoy about these trip reports is when they help convince other people to make the trip to their first conference. Hopefully by sharing their experience, it will convince you to come to the next conference: vBSDCon in Virginia, USA: Sept 7-9 EuroBSDCon in Paris, France: Sept 21-24 BSDTW in Taipei, Taiwan: November 11-12 (CFP ends July 31st) *** BSDCan 2017 - Trip report double-p (http://undeadly.org/cgi?action=article&sid=20170629150641) Prologue Most overheard in Tokyo was "see you in Ottawaaaaah", so with additional "personal item" being Groff I returned home to plan the trip to BSDCan. Dan was very helpful with getting all the preparations (immigration handling), thanks for that. Before I could start, I had to fix something: the handling of the goat. With a nicely created harness, I could just hang it along my backpack. Done that it went to the airport of Hamburg and check-in for an itinerary of HAM-MUC-YUL. While the feeder leg was a common thing, boarding to YUL was great - cabin-crew likes Groff :) Arriving in Montreal was like entering a Monsoon zone or something, sad! After the night the weather was still rain-ish but improving and i shuttled to Dorval VIARail station to take me to Ottawa (ever avoid AirCanada, right?). Train was late, but the conductor (or so) was nice to talk to - and wanted to know about Groff's facebook page :-P. Picking a cab in Ottawa to take me to "Residence" was easy at first - just that it was the wrong one. Actually my fault and so I had a "nice, short" walk to the actual one in the rain with wrong directions. Eventually I made it and after unpacking, refreshment it was time to hit the Goat BOF! Day 1 Since this was my first BSDCan I didnt exactly knew what to expect from this BOF. But it was like, we (Keeper, Dan, Allan, ..) would talk about "who's next" and things like that. How mistaken I was :). Besides the sheer amount of BSD people entering the not-so-yuuge Oak some Dexter sneaked in camouflage. The name-giver got a proper position to oversee the mess and I was glad I did not leave him behind after almost too many Creemores. Day 2 Something happened it's crystal blue on the "roof" and sun is trying its best to wake me up. To start the day, I pick breakfast at 'Father+Sons' - I can really recommend that. Very nice home made fries (almost hashbrowns) and fast delivery! Stuffed up I trott along to get to phessler's tutorial about BGP-for-sysadmins-and-developers. Peter did a great job, but the "lab" couldn't happen, since - oh surprise - the wifi was sluggish as hell. Must love the first day on a conference every time. Went to Hackroom in U90 afterwards, just to fix stuff "at home". IPsec giving pains again. Time to pick food+beer afterwards and since it's so easy to reach, we went to the Oak again. Having a nice backyard patio experience it was about time to meet new people. Cheers to Tom, Aaron, Nick, Philip and some more, we'd an awesome night there. I also invited some not-really-computer local I know by other means who was completly overwhelmed by what kind of "nerds" gather around BSD. He planned to stay "a beer" - and it was rather some more and six hours. Looks like "we" made some impression on him :). Day 3 Easy day, no tutorials at hand, so first picking up breakfast at F+S again and moving to hackroom in U90. Since I promised phessler to help with an localized lab-setup, I started to hack on a quick vagrant/ansible setup to mimic his BGP-lab and went quickly through most of it. Plus some more IPsec debugging and finally fixing it, we went early in the general direction of the Red Lion to pick our registration pack. But before that could happen it was called to have shawarma at 3brothers along. Given a tight hangover it wasn't the brightest idea to order a poutine m-(. Might be great the other day, it wasn't for me at the very time and had to throw away most of it :(. Eventually passing on to the Red Lion I made the next failure with just running into the pub - please stay at the front desk until "seated". I never get used to this concept. So after being "properly" seated, we take our beers and the registration can commence after we had half of it. So I register myself; btw it's a great idea to grant "not needed" stuff to charity. So dont pick "just because", think about it if you really need this or that gadget. Then I register Groff - he really needs badges - just to have Dru coming back to me some minutes later one to hand me the badge for Henning. That's just "amazing"; I dont know IF i want to break this vicious circle the other day, since it's so funny. Talked to Theo about the ongoing IPsec problems and he taught me about utrace(2) which looks "complicated" but might be an end of the story the other day. Also had a nice talk to Peter (H.) about some other ideas along books. BTW, did I pay for ongoing beers? I think Tom did - what a guy :). Arriving at the Residence, I had to find my bathroom door locked (special thing).. crazy thing is they dont have a master key at the venue, but to have to call in one from elsewhere. Short night shortened by another 30minutes :(. Day 4 Weather is improving into beach+sun levels - and it's Conference Day! The opening keynote from Geist was very interesting ("citation needed"). Afterwards I went to zfs-over-ssh, nothing really new (sorry Allan). But then Jason had a super interesting talk on how about to apply BSD for the health-care system in Australia. I hope I can help him with the last bits (rdomain!) in the end. While lunch I tried to recall my memories about utrace(2) while talking to Theo. Then it was about to present my talk and I think it was well perceipted. One "not so good" feedback was about not taking the audience more into account. I think I was asking every other five slides or so - but, well. The general feedback (in spoken terms) was quite good. I was a bit "confused" and I did likely a better job in Tokyo, but well. Happened we ended up in the Oak again.. thanks to mwl, shirkdog, sng, pitrh, kurtm for having me there :) Day 5 While the weather had to decide "what next", I rushed to the venue just to gather Reyk's talk about vmd(8). Afterwards it was MSTP from Paeps which was very interesting and we (OpenBSD) should look into it. Then happened BUG BOF and I invite all "coastal Germans" to cbug.de :) I had to run off for other reasons and came back to Dave's talk which was AWESOME. Following was Rod's talk.. well. While I see his case, that was very poor. The auction into closing was awesome again, and I spend $50 on a Tshirt. :) + Epilogue I totally got the exit dates wrong. So first cancel a booking of an Hotel and then rebook the train to YUL. So I have plenty of time "in the morning" to get breakfast with the local guy. After that he drives me to VIARail station and I dig into "business" cussions. Well, see you in Ottawa - or how about Paris, Taipei? Bind Broker (http://www.tedunangst.com/flak/post/bind-broker) Ted Unangst writes about an interesting idea he has He has a single big server, and lots of users who would like to share it, many want to run web servers. This would be great, but alas, archaic decisions made long ago mean that network sockets aren't really files and there's this weird concept of privileged ports. Maybe we could assign each user a virtual machine and let them do whatever they want, but that seems wasteful. Think of the megabytes! Maybe we could setup nginx.conf to proxy all incoming connections to a process of the user's choosing, but that only works for web sites and we want to be protocol neutral. Maybe we could use iptables, but nobody wants to do that. What we need is a bind broker. At some level, there needs to be some kind of broker that assigns IPs to users and resolves conflicts. It should be possible to build something of this nature given just the existing unix tools we have, instead of changing system design. Then we can deploy our broker to existing systems without upgrading or disrupting their ongoing operation. The bind broker watches a directory for the creation, by users, of unix domain sockets. Then it binds to the TCP port of the same name, and transfers traffic between them. A more complete problem specification is as follows. A top level directory, which contains subdirectories named after IP addresses. Each user is assigned a subdirectory, which they have write permission to. Inside each subdirectory, the user may create unix sockets named according to the port they wish to bind to. We might assign user alice the IP 10.0.0.5 and the user bob the IP 10.0.0.10. Then alice could run a webserver by binding to net/10.0.0.5/80 and bob could run a mail server by binding to net/10.0.0.10/25. This maps IP ownership (which doesn't really exist in unix) to the filesystem namespace (which does have working permissions). So this will be a bit different than jails. The idea is to use filesystem permissions to control which users can bind to which IP addresses and ports The broker is responsible for watching each directory. As new sockets are created, it should respond by binding to the appropriate port. When a socket is deleted, the network side socket should be closed as well. Whenever a connection is accepted on the network side, a matching connection is made on the unix side, and then traffic is copied across. A full set of example code is provided There's no completely portable way to watch a directory for changes. I'm using a kevent extension. Otherwise we might consider a timeout and polling with fstat, or another system specific interface (or an abstraction layer over such an interface). Otherwise, if one of our mappings is ready to read (accept), we have a new connection to handle. The first half is straightforward. We accept the connection and make a matching connect call to the unix side. Then I broke out the big cheat stick and just spliced the sockets together. In reality, we'd have to set up a read/copy/write loop for each end to copy traffic between them. That's not very interesting to read though. The full code, below, comes in at 232 lines according to wc. Minus includes, blank lines, and lines consisting of nothing but braces, it's 148 lines of stuff that actually gets executed by the computer. Add some error handling, and working read/write code, and 200 lines seems about right. A very interesting idea. I wonder about creating a virtual file system that would implement this and maybe do a bit more to fully flesh out this idea. What do you think? *** News Roundup Daemons and friendly Ninjas (https://euroquis.nl/bobulate/?p=1600) There's quite a lot of software that uses CMake as a (meta-)buildsystem. A quick count in the FreeBSD ports tree shows me 1110 ports (over a thousand) that use it. CMake generates buildsystem files which then direct the actual build — it doesn't do building itself. There are multiple buildsystem-backends available: in regular usage, CMake generates Makefiles (and does a reasonable job of producing Makefiles that work for GNU Make and for BSD Make). But it can generate Ninja, or Visual Studio, and other buildsystem files. It's quite flexible in this regard. Recently, the KDE-FreeBSD team has been working on Qt WebEngine, which is horrible. It contains a complete Chromium and who knows what else. Rebuilding it takes forever. But Tobias (KDE-FreeBSD) and Koos (GNOME-FreeBSD) noticed that building things with the Ninja backend was considerably faster for some packages (e.g. Qt WebEngine, and Evolution data-thingy). Tobias wanted to try to extend the build-time improvements to all of the CMake-based ports in FreeBSD, and over the past few days, this has been a success. Ports builds using CMake now default to using Ninja as buildsystem-backend. Here's a bitty table of build-times. These are one-off build times, so hardly scientifically accurate — but suggestive of a slight improvement in build time. Name Size GMake Ninja liblxt 50kB 0:32 0:31 llvm38 1655kB * 19:43 musescore 47590kB 4:00 3:54 webkit2-gtk3 14652kB 44:29 37:40 Or here's a much more thorough table of results from tcberner@, who did 5 builds of each with and without ninja. I've cut out the raw data, here are just the average-of-five results, showing usually a slight improvement in build time with Ninja. Name av make av ninj Delta D/Awo compiler-rt 00:08 00:07 -00:01 -14% openjpeg 00:06 00:07 +00:01 +17% marble 01:57 01:43 -00:14 -11% uhd 01:49 01:34 -00:15 -13% opencacscade 04:08 03:23 -00:45 -18% avidemux 03:01 02:49 -00:12 – 6% kdevelop 01:43 01:33 -00:10 – 9% ring-libclient 00:58 00:53 -00:05 – 8% Not everything builds properly with Ninja. This is usually due to missing dependencies that CMake does not discover; this shows up when foo depends on bar but no rule is generated for it. Depending on build order and speed, bar may be there already by the time foo gets around to being built. Doxygen showed this, where builds on 1 CPU core were all fine, but 8 cores would blow up occasionally. In many cases, we've gone and fixed the missing implicit dependencies in ports and upstreams. But some things are intractable, or just really need GNU Make. For this, the FreeBSD ports infrastructure now has a knob attached to CMake for switching a port build to GNU Make. Normal: USES=cmake Out-of-source: USES=cmake:outsource GNU Make: USES=cmake:noninja gmake OoS, GMake: USES=cmake:outsource,noninja gmake Bad: USES=cmake gmake For the majority of users, this has no effect, but for our package-building clusters, and for KDE-FreeBSD developers who build a lot of CMake-buildsystem software in a day it may add up to an extra coffee break. So I'll raise a shot of espresso to friendship between daemons and ninjas. Announcing the pkgsrc-2017Q2 release (http://mail-index.netbsd.org/pkgsrc-users/2017/07/10/msg025237.html) For the 2017Q2 release we welcome the following notable package additions and changes to the pkgsrc collection: Firefox 54 GCC 7.1 MATE 1.18 Ruby 2.4 Ruby on Rails 4.2 TeX Live 2017 Thunderbird 52.1 Xen 4.8 We say goodbye to: Ruby 1.8 Ruby 2.1 The following infrastructure changes were introduced: Implement optional new pkgtasks and init infrastructure for pkginstall. Various enhancements and fixes for building with ccache. Add support to USE_LANGUAGES for newer C++ standards. Enhanced support for SSP, FORTIFY, and RELRO. The GitHub mirror has migrated to https://github.com/NetBSD/pkgsrc In total, 210 packages were added, 43 packages were removed, and 1,780 package updates were processed since the pkgsrc-2017Q1 release. *** OpenBSD changes of note 624 (http://www.tedunangst.com/flak/post/openbsd-changes-of-note-624) There are a bunch, but here are a few that jump out: Start plugging some leaks. Compile kernels with umask 007. Install them minus read permissions. Pure preprocessor implementation of the roff .ec and .eo requests, though you are warned that very bad things will happen to anybody trying to use these macros in OpenBSD manuals. Random linking for arm64. And octeon. And alpha. And hppa. There's some variation by platform, because every architecture has the kernel loaded with different flavors of initial physical and virtual mappings. And landisk. And loongson. And sgi. And macppc. And a gap file for sparc64, but nobody yet dares split locore. And arm7. Errata for perl File::Path race condition. Some fixes for potential link attacks against cron. Add pledge violations to acct reporting. Take random linking to the next stage. More about KARL - kernel address randomized link. As noted, a few difficulties with hibernate and such, but the plan is coming together. Add a new function reorder_kernel() that relinks and installs the new kernel in the background on system startup. Add support for the bootblocks to detect hibernate and boot the previous kernel. Remove the poorly described “stuff” from ksh. Replace usage of TIOCSTI in csh using a more common IO loop. Kind of like the stuff in ksh, but part of the default command line editing and parsing code, csh would read too many characters, then send the ones it didn't like back into the terminal. Which is weird, right? Also, more importantly, eliminating the code that uses TIOCSTI to inject characters into ttys means that maybe TIOCSTI can be removed. Revamp some of the authentication logging in ssh. Add a verbose flag to rm so you can panic immediately upon seeing it delete the wrong file instead of waiting to discover your mistake after the fact. Update libexpat to version 2.2.1 which has some security fixes. Never trust an expat, that's my motto. Update inteldrm to code based on Linux 4.4.70. This brings us support for Skylake and Cherryview and better support for Broadwell and Valleyview. Also adds MST support. Fun times for people with newish laptops. *** OPNsense 17.1.9 released (https://opnsense.org/opnsense-17-1-9-released/) firewall: move gateway switching from system to firewall advanced settings firewall: keep category selection when changing tabs firewall: do not skip gateway switch parsing too early (contributed by Stephane Lesimple) interfaces: show VLAN description during edit firmware: opnsense-revert can now handle multiple packages at once firmware: opnsense-patch can now handle permission changes from patches dnsmasq: use canned –bogus-priv for noprivatereverse dnsmasq: separate log file, ACL and menu entries dynamic dns: fix update for IPv6 (contributed by Alexander Leisentritt) dynamic dns: remove usage of CURLAUTH_ANY (contributed by Alexander Leisentritt) intrusion detection: suppress “fast mode available” boot warning in PCAP mode openvpn: plugin framework adaption unbound: add local-zone type transparent for PTR zone (contributed by Davide Gerhard) unbound: separate log file, ACL and menu entries wizard: remove HTML from description strings mvc: group relation to something other than uuid if needed mvc: rework “item in” for our Volt templates lang: Czech to 100% translated (contributed by Pavel Borecki) plugins: zabbix-agent 1.1 (contributed by Frank Wall) plugins: haproxy 1.16 (contributed by Frank Wall) plugins: acme-client 1.8 (contributed by Frank Wall) plugins: tinc fix for switch mode (contributed by Johan Grip) plugins: monit 1.3 (contributed by Frank Brendel) src: support dhclient supersede statement for option 54 (contributed by Fabian Kurtz) src: add Intel Atom Cherryview SOC HSUART support src: add the ID for the Huawei ME909S LTE modem src: HardenedBSD Stack Clash mitigations[1] ports: sqlite 3.19.3[2] ports: openvpn 2.4.3[3] ports: sudo 1.8.20p2[4] ports: dnsmasq 2.77[5] ports: openldap 2.4.45[6] ports: php 7.0.20[7] ports: suricata 3.2.2[8] ports: squid 3.5.26[9] ports: carootnss 3.31 ports: bind 9.11.1-P2[10] ports: unbound 1.6.3[11] ports: curl 7.54.1[12] *** Beastie Bits Thinkpad x230 - trying to get TrackPoint / Touchpad working in X (http://lists.dragonflybsd.org/pipermail/users/2017-July/313519.html) FreeBSD deprecates all r-cmds (rcp, rlogin, etc.) (http://marc.info/?l=freebsd-commits-all&m=149918307723723&w=2) Bashfill - art for your terminal (https://max.io/bash.html) Go 1.9 release notes: NetBSD support is broken, please help (https://github.com/golang/go/commit/32002079083e533e11209824bd9e3a797169d1c4) Jest, A ReST api for creating and managing FreeBSD jails written in Go (https://github.com/altsrc-io/Jest) *** Feedback/Questions John - zfs send/receive (http://dpaste.com/3ANETHW#wrap) Callum - laptops (http://dpaste.com/11TV0BJ) & An update (http://dpaste.com/3A14BQ6#wrap) Lars - Snapshot of VM datadisk (http://dpaste.com/0MM37NA#wrap) Daryl - Jail managers (http://dpaste.com/0CDQ9EK#wrap) ***

time australia interview guide evolution german fun trip train hotels built tokyo whatsapp cheers id montreal pure norway ninjas picking ip remove developers rebuilding hamburg ottawa depending norwegian rod keeper residence implement cfp acl storage how to geist talked io btw open source python tutorials thin linux minus czech github enhanced arriving jest vm pep html roller install oak ips taipei bind cpu ports stuffed henning volt ssd revamp mst fs monsoons air canada unix chromium fortify browsing errata ssp royal oak digital ocean tcp visual studio ipv6 ptr groff peter h conference day bsd thickness bgp brokering compile red lion fcp bof freebsd vlan 2tb yul mbr pcap openbsd broadwell ipsec skylake 16gb ram cmake mstp netbsd allan jude ixsystems freebsd foundation trueos tarsnap bsdcan onion omega dragonflybsd vbsdcon

Episode 185: Skylake Goes for Gold… and Platinum; Intel Launches Xeon Scalable Platform

This Week in HPC

Play Episode Listen Later Jul 11, 2017 14:25

Addison Snell and Michael Feldman discuss the latest announcement from Intel.

gold platform intel launches platinum scalable xeon michael feldman skylake addison snell

Episode 181: New Systems for ARM, ShenWei and Skylake: Updates from ISC 17

This Week in HPC

Play Episode Listen Later Jun 21, 2017 9:41

Episode 181: New Systems for ARM, ShenWei and Skylake: Updates from ISC 17 by Intersect360 Research

CultCast #288 - iPad Pro and HomePod get great reviews!

Play Episode Listen Later Jun 16, 2017 59:54

This week: Reviews for the new iPad Pro and HomePod are bonkers Someone in the Apple supply chain just accidentally leaked some of iPhone 8’s marquee features Speed tests show a big improvement in the new MacBook Pro Apple finally offers us iCloud Storage sharing Our favorite unannounced iOS 11 features And we’ll tell you what we like (and don’t) about the apps and gadgets we’re currently testing in an all-new Under Review! This episode supported by Build a beautiful, responsive website quick at Squarespace.com. Enter offer code CultCast at checkout to get 10% off. Squarespace—Build it Beautiful. CultCloth will keep your iPhone 7, Apple Watch, Mac and iPad sparkling clean, and for a limited time you can use code CULTCAST to score a free CleanCloth with any order at CultCloth.co. Thanks to Kevin MacLeod at incompetech.com for the great music you hear on today's show. On the show this week @erfon / @bst3r / @lkahney Apple suppliers confirm two of iPhone 8’s biggest features http://www.cultofmac.com/487120/apple-supplier-will-ship-3d-facial-recognition-sensors-time-iphone-8/ Apple supplier Largan has confirmed it will be shipping its new 3D sensors in time for iPhone 8, while another has confirmed waterproofing and wireless charging technology. Largan’s sensors are capable of carrying out facial and iris recognition and provide an alternative to fingerprint scanning. They could be the solution Apple turns to if it is unable to embed a Touch ID scanner beneath the iPhone 8’s display. iPhone 7 and iPhone 7 Plus models are already splash- and water-resistant with an IP67 rating, but Apple's fine print warns that "splash, water, and dust resistance are not permanent conditions and resistance might decrease as a result of normal wear." iPhone water damage is not covered by Apple's warranties. But in Wednesday's shareholders meeting, [Robert] Hwang, [Wistron CEO] told reporters that the Assembly process for the previous generations of [iPhones] have not changed much, though new features like waterproof and wireless charging now require some different testing, and waterproof function will alter the assembly process a bit,” A report earlier this year said Apple's next iPhone models will feature improved IP68-rated water resistance (30 minutes in up to 3 feet of water) 2017 MacBook Pro is Up to 20% Faster Than Last Year's Model in Benchmarks https://www.macrumors.com/2017/06/07/2017-macbook-pro-benchmarks-geekbench/ Apple this week refreshed its MacBook Pro lineup with Intel's seventh-generation Kaby Lake processors, and early benchmarks for the notebooks suggest the 2017 models are up to 20 percent faster than the equivalent 2016 models equipped with Intel's sixth-generation Skylake processors. Only about a 30% improvement over 2012/2013 MacBook Pros... 10.5-inch iPad Pro review roundup: Apple’s most impressive tablet yet http://www.cultofmac.com/486616/10-5-inch-ipad-pro-review-roundup-apples-impressive-tablet-yet/ One of the big improvements of the new iPad Pro is its new variable refresh rate display, which Apple calls ProMotion, and means that it can run at 120Hz, or twice as fast as previous displays. It can also cycle down to 24Hz on still images. ARS has praised the new ProMotion by saying that it makes it easier to read text and scroll simultaneously, with the “ghosting” effect you get at 60Hz greatly reduced What everyone agrees on is that the new A10x processor is blazingly fast Smart keyboard is getting mixed reviews. Some saying it feels flimsy and unsatisfying to type on. Uses the newest Touch ID. Dramatically better antireflection coating. About 30% faster than 9.7inch iPad Pro, even more on multicore Geekbench tests I saw. Export times in Garage Band almost halved in some cases. Way better front (7MPX) and back cameras (12MPX), especially from the previous 12inch which had a 1.2 MPX camera. Fuller, better sounding external speakers iPad finally feels "pro" HomePod First Listen Impressions: 'Incredible' Audio With Bass-Heavy Sound That Easily Beats Echo https://www.macrumors.com/2017/06/06/homepod-first-impressions/ A few websites have now gotten the chance to hear how HomePod sounds in a quiet environment, and they're sharing their opinions online. Mashable: The audio wasn't just loud — filling a room with sound, good or bad, is easy — it was rich. Mashable noted that Apple had in-room comparisons with its competitors, including Sonos Play:3 and Amazon Echo, and that the HomePod easily beat the two other speakers in pure music playback quality. CNET: HomePod came off as bolder and more vivid than Sonos Play:3 in the [listening] experience, and a lot better than Amazon Echo. I'd also say the music sounded consistently vivid and crisp in a quiet space What HiFi: We also heard a pair of HomePods playing a live recording of Hotel California by The Eagles. The attention to detail was striking, with different instruments sounding discretely realised. Engadget called HomePod's audio "incredible,” if listening to the HomePod was like listening to a CD, then audio through the Echo sounded like AM radio. Apple Drops 2TB iCloud Storage Price to $9.99, Eliminates 1TB Option https://www.macrumors.com/2017/06/06/apple-updates-icloud-storage-prices/ Following the WWDC keynote, Apple updated and simplified its iCloud storage tiers 50GB: $0.99 200GB: $2.99 2TB: $9.99 (updated from 1TB) Apple is offering an option to share a 200GB or 2TB iCloud storage plan with family members, which is perhaps the reason behind the price drop. Google Photos offers free backup for photos and videos. Or you can back up via Plex premium. Offload unused apps Automatically offload unused apps if you’re low on storage. iOS 11 Preview: Control Center Gets Customizable With 3D Touch https://www.macrumors.com/2017/06/06/ios-11-preview-control-center/ Control center is now customizable, and you can use 3D touch to expand some control center icons (like the music control) iOS 11 Lets AirPods Users Change Tracks With a Double Tap https://www.macrumors.com/2017/06/07/ios-11-airpods-change-tracks-double-tap/ AirPods will soon be able to skip forwards and backwards between tracks with a double tap of either earpiece, thanks to a new additional setting in iOS 11. It’ll be possible to customize the AirPods so that a double tap on one earpiece skips, while double tapping on on the other earpiece goes to the previous track. Under Review Yuneec Typhoon H Lake Coloring App for iPhone and iPad Espin Bikes Espin bike deal Leander mentioned on Gilt

196: PostgreZFS

Play Episode Listen Later May 31, 2017 106:15

This week on BSD Now, we review the EuroBSDcon schedule, we explore the mysteries of Docker on OpenBSD, and show you how to run PostgreSQL on ZFS. This episode was brought to you by Headlines EuroBSDcon 2017 - Talks & Schedule published (https://2017.eurobsdcon.org/2017/05/26/talks-schedule-published/) The EuroBSDcon website was updated with the tutorial and talk schedule for the upcoming September conference in Paris, France. Tutorials on the 1st day: Kirk McKusick - An Introduction to the FreeBSD Open-Source Operating System, George Neville-Neil - DTrace for Developers, Taylor R Campbell - How to untangle your threads from a giant lock in a multiprocessor system Tutorials on the 2nd day: Kirk continues his Introduction lecture, Michael Lucas - Core concepts of ZFS (half day), Benedict Reuschling - Managing BSD systems with Ansible (half day), Peter Hessler - BGP for developers and sysadmins Talks include 3 keynotes (2 on the first day, beginning and end), another one at the end of the second day by Brendan Gregg Good mixture of talks of the various BSD projects Also, a good amount of new names and faces Check out the full talk schedule (https://2017.eurobsdcon.org/talks-schedule/). Registration is not open yet, but will be soon. *** OpenBSD on the Xiaomi Mi Air 12.5" (https://jcs.org/2017/05/22/xiaomiair) The Xiaomi Mi Air 12.5" (https://xiaomi-mi.com/notebooks/xiaomi-mi-notebook-air-125-silver/) is a basic fanless 12.5" Ultrabook with good build quality and decent hardware specs, especially for the money: while it can usually be had for about $600, I got mine for $489 shipped to the US during a sale about a month ago. Xiaomi offers this laptop in silver and gold. They also make a 13" version but it comes with an NVidia graphics chip. Since these laptops are only sold in China, they come with a Chinese language version of Windows 10 and only one or two distributors that carry them ship to the US. Unfortunately that also means they come with practically no warranty or support. Hardware > The Mi Air 12.5" has a fanless, 6th generation (Skylake) Intel Core m3 processor, 4Gb of soldered-on RAM, and a 128Gb SATA SSD (more on that later). It has a small footprint of 11.5" wide, 8" deep, and 0.5" thick, and weighs 2.3 pounds. > A single USB-C port on the right-hand side is used to charge the laptop and provide USB connectivity. A USB-C ethernet adapter I tried worked fine in OpenBSD. Whether intentional or not, a particular design touch I appreciated was that the USB-C port is placed directly to the right of the power button on the keyboard, so you don't have to look or feel around for the port when plugging in the power cable. > A single USB 3 type-A port is also available on the right side next to the USB-C port. A full-size HDMI port and a headphone jack are on the left-hand side. It has a soldered-on Intel 8260 wireless adapter and Bluetooth. The webcam in the screen bezel attaches internally over USB. > The chassis is all aluminum and has sufficient rigidity in the keyboard area. The 12.5" 1920x1080 glossy IPS screen has a fairly small bezel and while its hinge is properly weighted to allow opening the lid with one hand (if you care about that kind of thing), the screen does have a bit of top-end wobble when open, especially when typing on another laptop on the same desk. > The keyboard has a roomy layout and a nice clicky tactile with good travel. It is backlit, but with only one backlight level. When enabled via Fn+F10 (which is handled by the EC, so no OpenBSD support required), it will automatically shut off after not typing for a short while, automatically turning back once a key is pressed. Upgrades > An interesting feature of the Mi Air is that it comes with a 128Gb SATA SSD but also includes an open PCI-e slot ready to accept an NVMe SSD. > I upgraded mine with a Samsung PM961 256Gb NVMe SSD (left), and while it is possible to run with both drives in at the same time, I removed the Samsung CM871a 128Gb SATA (right) drive to save power. > The bottom case can be removed by removing the seven visible screws, in addition to the one under the foot in the middle back of the case, which just pries off. A spudger tool is needed to release all of the plastic attachment clips along the entire edge of the bottom cover. > Unfortunately this upgrade proved to be quite time consuming due to the combination of the limited UEFI firmware on the Mi Air and a bug in OpenBSD. A Detour into UEFI Firmware Variables > Unlike a traditional BIOS where one can boot into a menu and configure the boot order as well as enabling and disabling options such as "USB Hard Drive", the InsydeH2O UEFI firmware on the Xiaomi Air only provides the ability to adjust the boot order of existing devices. Any change or addition of boot devices must be done from the operating system, which is not possible under OpenBSD. > I booted to a USB key with OpenBSD on it and manually partitioned the new NVME SSD, then rsynced all of the data over from the old drive, but the laptop would not boot to the new NVME drive, instead showing an error message that there was no bootable OS. > Eventually I figured out that the GPT table that OpenBSD created on the NVMe disk was wrong due to a [one-off bug in the nvme driver](https://github.com/openbsd/src/commit/dc8298f669ea2d7e18c8a8efea509eed200cb989) which was causing the GPT table to be one sector too large, causing the backup GPT table to be written in the wrong location (and other utilities under Linux to write it over the OpenBSD area). I'm guessing the UEFI firmware would fail to read the bad GPT table on the disk that the boot variable pointed to, then declare that disk as missing, and then remove any variables that pointed to that disk. OpenBSD Support > The Mi Air's soldered-on Intel 8260 wireless adapter is supported by OpenBSD's iwm driver, including 802.11n support. The Intel sound chip is recognized by the azalia driver. > The Synaptics touchpad is connected via I2C, but is not yet supported. I am actively hacking on my dwiic driver to make this work and the touchpad will hopefully operate as a Windows Precision Touchpad via imt so I don't have to write an entirely new Synaptics driver. > Unfortunately since OpenBSD's inteldrm support that is ported from Linux is lagging quite a bit behind, there is no kernel support for Skylake and Kaby Lake video chips. Xorg works at 1920x1080 through efifb so the machine is at least usable, but X is not very fast and there is a noticeable delay when doing certain redrawing operations in xterm. Screen backlight can be adjusted through my OpenBSD port of intel_backlight. Since there is no hardware graphics support, this also means that suspend and resume do not work because nothing is available to re-POST the video after resume. Having to use efifb also makes it impossible to adjust the screen gamma, so for me, I can't use redshift for comfortable night-time hacking. Flaws > Especially taking into account the cheap price of the laptop, it's hard to find faults with the design. One minor gripe is that the edges of the case along the bottom are quite sharp, so when carrying the closed laptop, it can feel uncomfortable in one's hands. > While all of those things could be overlooked, unfortunately there is also a critical flaw in the rollover support in the keyboard/EC on the laptop. When typing certain combinations of keys quickly, such as holding Shift and typing "NULL", one's fingers may actually hold down the Shift, N, and U keys at the same time for a very brief moment before releasing N. Normally the keyboard/EC would recognize U being pressed after N is already down and send an interrupt for the U key. Unfortunately on this laptop, particular combinations of three keys do not interrupt for the third key at all until the second key is lifted, usually causing the third key not to register at all if typed quickly. I've been able to reproduce this problem in OpenBSD, Linux, and Windows, with the combinations of at least Shift+N+U and Shift+D+F. Holding Shift and typing the two characters in sequence quickly enough will usually fail to register the final character. Trying the combinations without Shift, using Control or Alt instead of Shift, or other character pairs does not trigger the problem. This might be a problem in the firmware on the Embedded Controller, or a defect in the keyboard circuitry itself. As I mentioned at the beginning, getting technical support for this machine is difficult because it's only sold in China. Docker on OpenBSD 6.1-current (https://medium.com/@dave_voutila/docker-on-openbsd-6-1-current-c620513b8110) Dave Voutila writes: So here's the thing. I'm normally a macOS user…all my hardware was designed in Cupertino, built in China. But I'm restless and have been toying with trying to switch my daily machine over to a non-macOS system sort of just for fun. I find Linux messy, FreeBSD not as Apple-laptop-friendly as it should be, and Windows a non-starter. Luckily, I found a friend in Puffy. Switching some of my Apple machines over to dual-boot OpenBSD left a gaping hole in my workflow. Luckily, all the hard work the OpenBSD team has done over the last year seems to have plugged it nicely! OpenBSD's hypervisor support officially made it into the 6.1 release, but after some experimentation it was rather time consuming and too fragile to get a Linux guest up and running (i.e. basically the per-requisite for Docker). Others had reported some success starting with QEMU and doing lots of tinkering, but after a wasted evening I figured I'd grab the latest OpenBSD snapshot and try what the openbsd-misc list suggested was improved Linux support in active development. 10 (11) Steps to docker are provided Step 0 — Install the latest OpenBSD 6.1 snapshot (-current) Step 1 — Configure VMM/VMD Step 2 — Grab an Alpine Linux ISO Step 3 — Make a new virtual disk image Step 4 — Boot Alpine's ISO Step 5 — Inhale that fresh Alpine air Step 6 — Boot Alpine for Reals Step 7 — Install Docker Step 8 — Make a User Step 9 — Ditch the Serial Console Step 10 — Test out your Docker instance I haven't done it yet, but I plan on installing docker-compose via Python's pip package manager. I prefer defining containers in the compose files. PostgreSQL + ZFS Best Practices and Standard Procedures (https://people.freebsd.org/~seanc/postgresql/scale15x-2017-postgresql_zfs_best_practices.pdf) Slides from Sean Chittenden's talk about PostgreSQL and ZFS at Scale 15x this spring Slides start with a good overview of Postgres and ZFS, and how to use them together To start, it walks through the basics of how PostgreSQL interacts with the filesystem (any filesystem) Then it shows the steps to take a good backup of PostgreSQL, then how to do it even better with ZFS Then an intro to ZFS, and how Copy-on-Write changes host PostgreSQL interacts with the filesystem Overview of how ZFS works ZFS Tuning tips: Compression, Recordsize, atime, when to use mostly ARC vs sharedbuffer, plus pgrepack Followed by a discussion of the reliability of SSDs, and their Bit Error Rate (BER) A good SSD has a 4%/year chance of returning the wrong data. A cheap SSD 34% If you put 20 SSDs in a database server, that means 58% (Good SSDs) to 99.975% (Lowest quality commercially viable SSD) chance of an error per year Luckily, ZFS can detect and correct these errors This applies to all storage, not just SSDs, every device fails More Advice: Use quotas and reservations to avoid running out of space Schedule Periodic Scrubs One dataset per database Backups: Live demo of rm -rf'ing the database and getting it back Using clones to test upgrades on real data Naming Conventions: Use a short prefix not on the root filesystem (e.g. /db) Encode the PostgreSQL major version into the dataset name Give each PostgreSQL cluster its own dataset (e.g. pgdb01) Optional but recommended: one database per cluster Optional but recommended: one app per database Optional but recommended: encode environment into DB name Optional but recommended: encode environment into DB username using ZFS Replication Check out the full detailed PDF and implement a similar setup for your database needs *** News Roundup TrueOS Evolving Its "Stable" Release Cycle (https://www.trueos.org/blog/housekeeping-update-infrastructure-trueos-changes/) TrueOS is reformulating its Stable branch based on feedback from users. The goal is to have a “release” of the stable branch every 6 months, for those who do not want to live on the edge with the rapid updates of the full rolling release Most of the TrueOS developers work for iX Systems in their Tennessee office. Last month, the Tennessee office was moved to a different location across town. As part of the move, we need to move all our servers. We're still getting some of the infrastructure sorted before moving the servers, so please bear with us as we continue this process. As we've continued working on TrueOS, we've heard a significant portion of the community asking for a more stable “STABLE” release of TrueOS, maybe something akin to an old PC-BSD version release. In order to meet that need, we're redefining the TrueOS STABLE branch a bit. STABLE releases are now expected to follow a six month schedule, with more testing and lots of polish between releases. This gives users the option to step back a little from the “cutting edge” of development, but still enjoy many of the benefits of the “rolling release” style and the useful elements of FreeBSD Current. Critical updates like emergency patches and utility bug fixes are still expected to be pushed to STABLE on a case-by-case basis, but again with more testing and polish. This also applies to version updates of the Lumina and SysAdm projects. New, released work from those projects will be tested and added to STABLE outside the 6 month window as well. The UNSTABLE branch continues to be our experimental “cutting edge” track, and users who want to follow along with our development and help us or FreeBSD test new features are still encouraged to follow the UNSTABLE track by checking that setting in their TrueOS Update Manager. With boot environments, it will be easy to switch back and forth, so you can have the best of both worlds. Use the latest bleeding edge features, but knowing you can fall back to the stable branch with just a reboot As TrueOS evolves, it is becoming clearer that one role of the system is to function as a “test platform” for FreeBSD. In order to better serve this role, TrueOS will support both OpenRC and the FreeBSD RC init systems, giving users the choice to use either system. While the full functionality isn't quite ready for the next STABLE update, it is planned for addition after the last bit of work and testing is complete. Stay tuned for an upcoming blog post with all the details of this change, along with instructions how to switch between RC and OpenRC. This is the most important change for me. I used TrueOS as an easy way to run the latest version of -CURRENT on my laptop, to use it as a user, but also to do development. When TrueOS deviates from FreeBSD too much, it lessens the power of my expertise, and complicates development and debugging. Being able to switch back to RC, even if it takes another minute to boot, will bring TrueOS back to being FreeBSD + GUI and more by default, instead of a science project. We need both of those things, so having the option, while more work for the TrueOS team, I think will be better for the entire community *** Logical Domains on SunFire T2000 with OpenBSD/sparc64 (http://www.h-i-r.net/2017/05/logical-domains-on-sunfire-t2000-with.html) A couple of years ago, I picked up a Sun Fire T2000. This is a 2U rack mount server. Mine came with four 146GB SAS drives, a 32-core UltraSPARC T1 CPU and 32GB of RAM. Sun Microsystems incorporated Logical Domains (LDOMs) on this class of hardware. You don't often need 32 threads and 32GB of RAM in a single server. LDOMs are a kind of virtualization technology that's a bit closer to bare metal than vmm, Hyper-V, VirtualBox or even Xen. It works a bit like Xen, though. You can allocate processor, memory, storage and other resources to virtual servers on-board, with a blend of firmware that supports the hardware allocation, and some software in userland (on the so-called primary or control domain, similar to Xen DomU) to control it. LDOMs are similar to what IBM calls Logical Partitions (LPARs) on its Mainframe and POWER series computers. My day job from 2006-2010 involved working with both of these virtualization technologies, and I've kind of missed it. While upgrading OpenBSD to 6.1 on my T2000, I decided to delve into LDOM support under OpenBSD. This was pretty easy to do, but let's walk through it Resources: The ldomctl(8) man page (http://man.openbsd.org/OpenBSD-current/man8/sparc64/ldomctl.8) tedu@'s write-up on Flak (for a different class of server) (http://www.tedunangst.com/flak/post/OpenBSD-on-a-Sun-T5120) A Google+ post by bmercer@ (https://plus.google.com/101694200911870273983/posts/jWh4rMKVq97) Once you get comfortable with the fact that there's a little-tiny computer (the ALOM) powered by VXWorks inside that's acting as the management system and console (there's no screen or keyboard/mouse input), Installing OpenBSD on the base server is pretty straightforward. The serial console is an RJ-45 jack, and, yes, the ubiquitous blue-colored serial console cables you find for certain kinds of popular routers will work fine. OpenBSD installs quite easily, with the same installer you find on amd64 and i386. I chose to install to /dev/sd0, the first SAS drive only, leaving the others unused. It's possible to set them up in a hardware RAID configuration using tools available only under Solaris, or use softraid(4) on OpenBSD, but I didn't do this. I set up the primary LDOM to use the first ethernet port, em0. I decided I wanted to bridge the logical domains to the second ethernet port. You could also use a bridge and vether interface, with pf and dhcpd to create a NAT environment, similar to how I networked the vmm(4) systems. Create an LDOM configuration file. You can put this anywhere that's convenient. All of this stuff was in a "vm" subdirectory of my home. I called it ldom.conf: domain primary { vcpu 8 memory 8G } domain puffy { vcpu 8 memory 4G vdisk "/home/axon/vm/ldom1" vnet } Make as many disk images as you want, and make as many additional domain clauses as you wish. Be mindful of system resources. I couldn't actually allocate a full 32GB of RAM across all the LDOMs I eventually provisioned seven LDOMs (in addition to the primary) on the T2000, each with 3GB of RAM and 4 vcpu cores. If you get creative with use of network interfaces, virtual ethernet, bridges and pf rules, you can run a pretty complex environment on a single chassis, with services that are only exposed to other VMs, a DMZ segment, and the internal LAN. A nice tutorial, and an interesting look at an alternative platform that was ahead of its time *** documentation is thoroughly hard (http://www.tedunangst.com/flak/post/documentation-is-thoroughly-hard) Ted Unangst has a new post this week about documentation: Documentation is good, so therefore more documentation must be better, right? A few examples where things may have gotten out of control A fine example is the old OpenBSD install instructions. Once you've installed OpenBSD once or twice, the process is quite simple, but you'd never know this based on reading the instructions. Compare the files for 4.8 INSTALL and 5.8 INSTALL. Both begin with a brief intro to the project. Then 4.8 has an enormous list of mirrors, which seems fairly redundant if you've already found the install file. Followed by an enormous list of every supported variant of every supported device. Including a table of IO port configurations for ISA devices. Finally, after 1600 lines of introduction we get to the actual installation instructions. (Compared to line 231 for 5.8.) This includes a full page of text about how to install from tape, which nobody ever does. It took some time to recognize that all this documentation was actually an impediment to new users. Attempting to answer every possible question floods the reader with information for questions they were never planning to ask. Part of the problem is how the information is organized. Theoretically it makes sense to list supported hardware before instructions. After all, you can't install anything if it's not supported, right? I'm sure that was considered when the device list was originally inserted above the install instructions. But as a practical matter, consulting a device list is neither the easiest nor fastest way to determine what actually works. In the FreeBSD docs tree, we have been doing a facelift project, trying to add ‘quick start' sections to each chapter to let you get to the more important information first. It is also helpful to move data in the forms of lists and tables to appendices or similar, where they can easily be references, but are not blocking your way to the information you are actually hunting for An example of nerdview signage (http://languagelog.ldc.upenn.edu/nll/?p=29866). “They have in effect provided a sign that will tell you exactly what the question is provided you can already supply the answer.” That is, the logical minds of technical people often decide to order information in an order that makes sense to them, rather than in the order that will be most useful to the reader In the end, I think “copy diskimage to USB and follow prompts” is all the instructions one should need, but it's hard to overcome the unease of actually making the jump. What if somebody is confused or uncertain? Why is this paragraph more redundant than that paragraph? (And if we delete both, are we cutting too much?) Sometimes we don't need to delete the information. Just hide it. The instructions to upgrade to 4.8 and upgrade to 5.8 are very similar, with a few differences because every release is a little bit different. The pages look very different, however, because the not at all recommended kernel free procedure, which takes up half the page, has been hidden from view behind some javascript and only expanded on demand. A casual browser will find the page and figure the upgrade process will be easy, as opposed to some long ordeal. This is important as well, it was my original motivation for working on the FreeBSD Handbook's ZFS chapter. The very first section of the chapter was the custom kernel configuration required to run ZFS on i386. That scared many users away. I moved that to the very end, and started with why you might want to use ZFS. Much more approachable. Sometimes it's just a tiny detail that's overspecified. The apmd manual used to explain exactly which CPU idle time thresholds were used to adjust frequency. Those parameters, and the algorithm itself, were adjusted occasionally in response to user feedback, but sometimes the man page lagged behind. The numbers are of no use to a user. They're not adjustable without recompiling. Knowing that the frequency would be reduced at 85% idle vs 90% idle doesn't really offer much guidance as to whether to enable auto scaling or not. Deleting this detail ensured the man page was always correct and spares the user the cognitive load of trying to solve an unnecessary math problem. For fun: For another humorous example, it was recently observed that the deja-dup package provides man page translations for Australia, Canada, and Great Britain. I checked, the pages are in fact not quite identical. Some contain typo fixes that didn't propagate to other translations. Project idea: attempt to identify which country has the most users, or most fastidious users, by bug fixes to localized man pages. lldb on BeagleBone Black (https://lists.freebsd.org/pipermail/freebsd-arm/2017-May/016260.html) I reliably managed to build (lldb + clang/lld) from the svn trunk of LLVM 5.0.0 on my Beaglebone Black running the latest snapshot (May 20th) of FreeBSD 12.0-CURRENT, and the lldb is working very well, and this includes single stepping and ncurses-GUI mode, while single stepping with the latest lldb 4.0.1 from the ports does not work. In order to reliably build LLVM 5.0.0 (svn), I set up a 1 GB swap partition for the BBB on a NFSv4 share on a FreeBSD fileserver in my network - I put a howto of the procedure on my BLog: https://obsigna.net/?p=659 The prerequesites on the Beaglebone are: ``` pkg install tmux pkg install cmake pkg install python pkg install libxml2 pkg install swig30 pkg install ninja pkg install subversion ``` On the FreeBSD fileserver: ``` /pathtothe/bbb_share svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm cd llvm/tools svn co http://llvm.org/svn/llvm-project/cfe/trunk clang svn co http://llvm.org/svn/llvm-project/lld/trunk lld svn co http://llvm.org/svn/llvm-project/lldb/trunk lldb ``` + On the Beaglebone Black: # mount_nfs -o noatime,readahead=4,intr,soft,nfsv4 server:/path_to_the/bbb_share /mnt # cd /mnt # mkdir build # cmake -DLLVM_TARGETS_TO_BUILD="ARM" -DCMAKE_BUILD_TYPE="MinSizeRel" -DLLVM_PARALLEL_COMPILE_JOBS="1" -DLLVM_PARALLEL_LINK_JOBS="1" -G Ninja .. I execute the actual build command from within a tmux session, so I may disconnect during the quite long (40 h) build: ``` tmux new "ninja lldb install" ``` When debugging in GUI mode using the newly build lldb 5.0.0-svn, I see only a minor issue, namely UTF8 strings are not displayed correctly. This happens in the ncurses-GUI only, and this is an ARM issue, since it does not occur on x86 machines. Perhaps this might be related to the signed/unsigned char mismatch between ARM and x86. Beastie Bits Triangle BSD Meetup on June 27th (https://www.meetup.com/Triangle-BSD-Users-Group/events/240247251/) Support for Controller Area Networks (CAN) in NetBSD (http://www.feyrer.de/NetBSD/bx/blosxom.cgi/nb_20170521_0113.html) Notes from Monday's meeting (http://mailman.uk.freebsd.org/pipermail/ukfreebsd/2017-May/014104.html) RunBSD - A site about the BSD family of operating systems (http://runbsd.info/) BSDCam(bridge) 2017 Travel Grant Application Now Open (https://www.freebsdfoundation.org/blog/bsdcam-2017-travel-grant-application-now-open/) New BSDMag has been released (https://bsdmag.org/download/nearly-online-zpool-switching-two-freebsd-machines/) *** Feedback/Questions Philipp - A show about byhve (http://dpaste.com/390F9JN#wrap) Jake - byhve Support on AMD (http://dpaste.com/0DYG5BD#wrap) CY - Pledge and Capsicum (http://dpaste.com/1YVBT12#wrap) CY - OpenSSL relicense Issue (http://dpaste.com/3RSYV23#wrap) Andy - Laptops (http://dpaste.com/0MM09EX#wrap) ***

canada australia power china apple interview guide france project chinese tennessee write current blog shift built os mine scale windows ibm developers intel ram screen switching copy ditch compare talks arm great britain arc alt raid nvidia hardware storage how to io bluetooth usb open source gb python stable tutorials gpt rj linux upgrades flaws lan db sas bbb amd macos lowest optional documentation detour slides install ips cpu rc ec xiaomi 4g ssd unstable usb c docker bios compression solaris deleting cupertino inhale hdmi puffy dmz pci theoretically sun microsystems vms digital ocean mainframe postgresql lumina bsd ansible postgres nvme flak xen freebsd 2u uefi virtualbox 32gb 4gb encode llvm hyper v zfs capsicum alom ultrabooks openbsd 3gb qemu nvme ssd skylake i2c kaby lake synaptics netbsd xorg allan jude ixsystems beaglebone black tarsnap beaglebone trueos bsd now pc bsd eurobsdcon dragonflybsd openrc

CultCast #284 - New MacBook Pros at WWDC!

Inside the Datacenter - Connected Social Media

Play Episode Listen Later May 19, 2017 50:25

This week: new MacBook Pros at WWDC? Insiders say yes! We’ll tell you all we know. Plus: why future Macs are about to get much faster CPUs; Apple makes a big move to bring manufacturing back to the US; and we’ll wrap up with 5 weird and whacky facts about the new Apple Park campus. This episode supported by Build a beautiful, responsive website quick at Squarespace.com. Enter offer code CultCast at checkout to get 10% off. Squarespace—Build it Beautiful. CultCloth will keep your iPhone 7, Apple Watch, Mac and iPad sparkling clean, and for a limited time you can use code CULTCAST to score a free CleanCloth with any order at CultCloth.co. We also want to give Kevin MacLeod at incompetech.com a thanks for the great music you hear on today's show. On the show this week @erfon / @bst3r / @lkahney Apple may reveal three new MacBooks at WWDC http://www.cultofmac.com/481877/new-macbooks-wwdc/ Apple hasn’t released new hardware at a WWDC keynote since 2013, but the company is allegedly planning to unveil a new lineup of MacBooks, according to a report that claims the new machines will pack Intel’s new Kaby Lake processor to bring more speed than ever. Three new laptops will debut at WWDC 2017, claims Bloomberg, citing “people familiar with the matter.” Both the MacBook Pro and 12-inch MacBook will be updated with new Intel chips. Apple is also supposedly considering updating the 13-inch MacBook Air with a new processor, too, which would be quite a surprise as most observers assumed the machine was on its last legs now that the MacBook and MacBook Pro are thinner. Sales of the old MacBook Air remain “surprisingly strong” due to its cheap price tag, claims one of the report’s sources. What features needed to make the machine exciting again Shows Apple may be getting about making Mac great again. Intel: Cannonlake CPUs will be more than 15 percent faster than Kaby Lake http://www.pcworld.com/article/3167942/components-processors/intel-cannonlake-will-be-more-than-15-percent-faster-than-kaby-lake.html Meager performance gains aren’t all Apple’s fault Chipmakers in past years focused on increasing performance by raising the clock frequency. But that made chips power hungry, and their focus shifted to adding cores, which boosted performance but also added battery life to laptops. Then the focus turned to integrating technologies like graphics and I/O buses inside processors. Gaming and virtual reality have brought a focus back to raw CPU performance. The performance improvements from Skylake to Kaby Lake topped out at 15 percent. The CPU performance boost for Cannonlake should be at least that, Intel said. The gaming market is exploding, especially eSports, and demand for high-performance Core i7 chips skyrocketed last year Intel may be trying to catch up with AMD, which is boasting a 40 percent performance improvement for its upcoming Ryzen chips. Apple’s standalone Siri could look a lot like Echo Show http://www.cultofmac.com/481625/apples-standalone-siri-look-lot-like-echo-show/ Apple’s Amazon Echo rival standalone Siri speaker will come with a touch-sensitive display, claims KGI Securities analyst Ming-Chi Kuo. In a note to clients over the weekend, Kuo suggested that the Siri home speaker will have a “touch panel,” although it’s not known whether this will be a full-on screen or a simpler touch-based interface of some sort. Apple gives $200 million to iPhone glassmaker to promote U.S. manufacturing http://www.cultofmac.com/481305/apple-gives-200-million-iphone-glassmakers-promote-u-s-manufacturing/ You may have heard that Tim Cook recently announced a $1 billion Advanced Manufacturing Fund in an interview with Mad Money’s Jim Cramer at Apple Campus, he said the goal was to push “people to do advanced manufacturing in the United States.” Rather than pulling from its enormous pile of overseas cash, Apple is borrowing the money for its $1 billion fund since that is cheaper than paying to repatriate its foreign money pile. Apple has awarded Corning the first grant of its $1 billion investment aimed at boosting high-tech manufacturing jobs in the United States. The glassmaker will receive $200 million. Apple’s contribution is part of its “Advanced Manufacturing Fund” will support Corning’s R&D, capital equipment needs, and state-of-the-art glass processing. Apple Park originally looked like a penis and 5 other wild facts http://www.cultofmac.com/481821/apple-park-penis-shaped-campus/ How Apple Park campus almost looked like a giant dong https://www.wired.com/2015/10/look-apples-newest-spaceship-building/

Re-Architecting the Intel Xeon Processor Family for Scalability – Intel Chip Chat – Episode 529

Play Episode Listen Later May 8, 2017

In this Intel Chip Chat audio podcast with Allyson Klein: Jennifer Huffstetler, Director of Datacenter Product Marketing at Intel, joins us to announce a new platform brand for Intel’s datacenter processors. Cued by the significant architectural leaps in what many have heard of as “Skylake,” Intel has re-architected the Intel Xeon processor family brand to […]

188: And then the murders began

Play Episode Listen Later Apr 5, 2017 83:39

Today on BSD Now, the latest Dragonfly BSD release, RaidZ performance, another OpenSSL Vulnerability, and more; all this week on BSD Now. This episode was brought to you by Headlines DragonFly BSD 4.8 is released (https://www.dragonflybsd.org/release48/) Improved kernel performance This release further localizes cache lines and reduces/removes cache ping-ponging on globals. For bulk builds on many-cores or multi-socket systems, we have around a 5% improvement, and certain subsystems such as namecache lookups and exec()s see massive focused improvements. See the corresponding mailing list post with details. Support for eMMC booting, and mobile and high-performance PCIe SSDs This kernel release includes support for eMMC storage as the boot device. We also sport a brand new SMP-friendly, high-performance NVMe SSD driver (PCIe SSD storage). Initial device test results are available. EFI support The installer can now create an EFI or legacy installation. Numerous adjustments have been made to userland utilities and the kernel to support EFI as a mainstream boot environment. The /boot filesystem may now be placed either in its own GPT slice, or in a DragonFly disklabel inside a GPT slice. DragonFly, by default, creates a GPT slice for all of DragonFly and places a DragonFly disklabel inside it with all the standard DFly partitions, such that the disk names are roughly the same as they would be in a legacy system. Improved graphics support The i915 driver has been updated to match the version found with the Linux 4.6 kernel. Broadwell and Skylake processor users will see improvements. Other user-affecting changes Kernel is now built using -O2. VKernels now use COW, so multiple vkernels can share one disk image. powerd() is now sensitive to time and temperature changes. Non-boot-filesystem kernel modules can be loaded in rc.conf instead of loader.conf. *** #8005 poor performance of 1MB writes on certain RAID-Z configurations (https://github.com/openzfs/openzfs/pull/321) Matt Ahrens posts a new patch for OpenZFS Background: RAID-Z requires that space be allocated in multiples of P+1 sectors,because this is the minimum size block that can have the required amount of parity. Thus blocks on RAIDZ1 must be allocated in a multiple of 2 sectors; on RAIDZ2 multiple of 3; and on RAIDZ3 multiple of 4. A sector is a unit of 2^ashift bytes, typically 512B or 4KB. To satisfy this constraint, the allocation size is rounded up to the proper multiple, resulting in up to 3 "pad sectors" at the end of some blocks. The contents of these pad sectors are not used, so we do not need to read or write these sectors. However, some storage hardware performs much worse (around 1/2 as fast) on mostly-contiguous writes when there are small gaps of non-overwritten data between the writes. Therefore, ZFS creates "optional" zio's when writing RAID-Z blocks that include pad sectors. If writing a pad sector will fill the gap between two (required) writes, we will issue the optional zio, thus doubling performance. The gap-filling performance improvement was introduced in July 2009. Writing the optional zio is done by the io aggregation code in vdevqueue.c. The problem is that it is also subject to the limit on the size of aggregate writes, zfsvdevaggregationlimit, which is by default 128KB. For a given block, if the amount of data plus padding written to a leaf device exceeds zfsvdevaggregation_limit, the optional zio will not be written, resulting in a ~2x performance degradation. The solution is to aggregate optional zio's regardless of the aggregation size limit. As you can see from the graphs, this can make a large difference in performance. I encourage you to read the entire commit message, it is well written and very detailed. *** Can you spot the OpenSSL vulnerability (https://guidovranken.wordpress.com/2017/01/28/can-you-spot-the-vulnerability/) This code was introduced in OpenSSL 1.1.0d, which was released a couple of days ago. This is in the server SSL code, ssl/statem/statemsrvr.c, sslbytestocipherlist()), and can easily be reached remotely. Can you spot the vulnerability? So there is a loop, and within that loop we have an ‘if' statement, that tests a number of conditions. If any of those conditions fail, OPENSSLfree(raw) is called. But raw isn't the address that was allocated; raw is increment every loop. Hence, there is a remote invalid free vulnerability. But not quite. None of those checks in the ‘if' statement can actually fail; earlier on in the function, there is a check that verifies that the packet contains at least 1 byte, so PACKETget1 cannot fail. Furthermore, earlier in the function it is verified that the packet length is a multiple of 3, hence PACKETcopybytes and PACKET_forward cannot fail. So, does the code do what the original author thought, or expected it to do? But what about the next person that modifies that code, maybe changing or removing one of the earlier checks, allowing one of those if conditions to fail, and execute the bad code? Nonetheless OpenSSL has acknowledged that the OPENSSL_free line needs a rewrite: Pull Request #2312 (https://github.com/openssl/openssl/pull/2312) PS I'm not posting this to ridicule the OpenSSL project or their programming skills. I just like reading code and finding corner cases that impact security, which is an effort that ultimately works in everybody's best interest, and I like to share what I find. Programming is a very difficult enterprise and everybody makes mistakes. Thanks to Guido Vranken for the sharp eye and the blog post *** Research Debt (http://distill.pub/2017/research-debt/) I found this article interesting as it relates to not just research, but a lot of technical areas in general Achieving a research-level understanding of most topics is like climbing a mountain. Aspiring researchers must struggle to understand vast bodies of work that came before them, to learn techniques, and to gain intuition. Upon reaching the top, the new researcher begins doing novel work, throwing new stones onto the top of the mountain and making it a little taller for whoever comes next. People expect the climb to be hard. It reflects the tremendous progress and cumulative effort that's gone into the research. The climb is seen as an intellectual pilgrimage, the labor a rite of passage. But the climb could be massively easier. It's entirely possible to build paths and staircases into these mountains. The climb isn't something to be proud of. The climb isn't progress: the climb is a mountain of debt. Programmers talk about technical debt: there are ways to write software that are faster in the short run but problematic in the long run. Poor Exposition – Often, there is no good explanation of important ideas and one has to struggle to understand them. This problem is so pervasive that we take it for granted and don't appreciate how much better things could be. Undigested Ideas – Most ideas start off rough and hard to understand. They become radically easier as we polish them, developing the right analogies, language, and ways of thinking. Bad abstractions and notation – Abstractions and notation are the user interface of research, shaping how we think and communicate. Unfortunately, we often get stuck with the first formalisms to develop even when they're bad. For example, an object with extra electrons is negative, and pi is wrong Noise – Being a researcher is like standing in the middle of a construction site. Countless papers scream for your attention and there's no easy way to filter or summarize them. We think noise is the main way experts experience research debt. There's a tradeoff between the energy put into explaining an idea, and the energy needed to understand it. On one extreme, the explainer can painstakingly craft a beautiful explanation, leading their audience to understanding without even realizing it could have been difficult. On the other extreme, the explainer can do the absolute minimum and abandon their audience to struggle. This energy is called interpretive labor Research distillation is the opposite of research debt. It can be incredibly satisfying, combining deep scientific understanding, empathy, and design to do justice to our research and lay bare beautiful insights. Distillation is also hard. It's tempting to think of explaining an idea as just putting a layer of polish on it, but good explanations often involve transforming the idea. This kind of refinement of an idea can take just as much effort and deep understanding as the initial discovery. + The distillation can often times require an entirely different set of skills than the original creation of the idea. Almost all of the BSD projects have some great ideas or subsystems that just need distillation into easy to understand and use platforms or tools. Like the theoretician, the experimentalist or the research engineer, the research distiller is an integral role for a healthy research community. Right now, almost no one is filling it. Anyway, if that bit piqued your interest, go read the full article and the suggested further reading. *** News Roundup And then the murders began. (https://blather.michaelwlucas.com/archives/2902) A whole bunch of people have pointed me at articles like this one (http://thehookmag.com/2017/03/adding-murders-began-second-sentence-book-makes-instantly-better-125462/), which claim that you can improve almost any book by making the second sentence “And then the murders began.” It's entirely possible they're correct. But let's check, with a sampling of books. As different books come in different tenses and have different voices, I've made some minor changes. “Welcome to Cisco Routers for the Desperate! And then the murders begin.” — Cisco Routers for the Desperate, 2nd ed “Over the last ten years, OpenSSH has become the standard tool for remote management of Unix-like systems and many network devices. And then the murders began.” — SSH Mastery “The Z File System, or ZFS, is a complicated beast, but it is also the most powerful tool in a sysadmin's Batman-esque utility belt. And then the murders begin.” — FreeBSD Mastery: Advanced ZFS “Blood shall rain from the sky, and great shall be the lamentation of the Linux fans. And then, the murders will begin.” — Absolute FreeBSD, 3rd Ed Netdata now supports FreeBSD (https://github.com/firehol/netdata) netdata is a system for distributed real-time performance and health monitoring. It provides unparalleled insights, in real-time, of everything happening on the system it runs (including applications such as web and database servers), using modern interactive web dashboards. From the release notes: apps.plugin ported for FreeBSD Check out their demo sites (https://github.com/firehol/netdata/wiki) *** Distrowatch Weekly reviews RaspBSD (https://distrowatch.com/weekly.php?issue=20170220#raspbsd) RaspBSD is a FreeBSD-based project which strives to create a custom build of FreeBSD for single board and hobbyist computers. RaspBSD takes a recent snapshot of FreeBSD and adds on additional components, such as the LXDE desktop and a few graphical applications. The RaspBSD project currently has live images for Raspberry Pi devices, the Banana Pi, Pine64 and BeagleBone Black & Green computers. The default RaspBSD system is quite minimal, running a mere 16 processes when I was logged in. In the background the operating system runs cron, OpenSSH, syslog and the powerd power management service. Other than the user's shell and terminals, nothing else is running. This means RaspBSD uses little memory, requiring just 16MB of active memory and 31MB of wired or kernel memory. I made note of a few practical differences between running RaspBSD on the Pi verses my usual Raspbian operating system. One minor difference is RaspBSD turns off the Pi's external power light after booting. Raspbian leaves the light on. This means it looks like the Pi is off when it is running RaspBSD, but it also saves a little electricity. Conclusions: Apart from these little differences, running RaspBSD on the Pi was a very similar experience to running Raspbian and my time with the operating system was pleasantly trouble-free. Long-term, I think applying source updates to the base system might be tedious and SD disk operations were slow. However, the Pi usually is not utilized for its speed, but rather its low cost and low-energy usage. For people who are looking for a small home server or very minimal desktop box, RaspBSD running on the Pi should be suitable. Research UNIX V8, V9 and V10 made public by Alcatel-Lucent (https://media-bell-labs-com.s3.amazonaws.com/pages/20170327_1602/statement%20regarding%20Unix%203-7-17.pdf) Alcatel-Lucent USA Inc. (“ALU-USA”), on behalf of itself and Nokia Bell Laboratories agrees, to the extent of its ability to do so, that it will not assert its copyright rights with respect to any non-commercial copying, distribution, performance, display or creation of derivative works of Research Unix®1 Editions 8, 9, and 10. Research Unix is a term used to refer to versions of the Unix operating system for DEC PDP-7, PDP-11, VAX and Interdata 7/32 and 8/32 computers, developed in the Bell Labs Computing Science Research Center. The version breakdown can be viewed on its Wikipedia page (https://en.wikipedia.org/wiki/Research_Unix) It only took 30+ years, but now they're public You can grab them from here (http://www.tuhs.org/Archive/Distributions/Research/) If you're wondering what happened with Research Unix, After Version 10, Unix development at Bell Labs was stopped in favor of a successor system, Plan 9 (http://plan9.bell-labs.com/plan9/); which itself was succeeded by Inferno (http://www.vitanuova.com/inferno/). *** Beastie Bits The BSD Family Tree (https://github.com/freebsd/freebsd/blob/master/share/misc/bsd-family-tree) Unix Permissions Calculator (http://permissions-calculator.org/) NAS4Free release 11.0.0.4 now available (https://sourceforge.net/projects/nas4free/files/NAS4Free-11.0.0.4/11.0.0.4.4141/) Another BSD Mag released for free downloads (https://bsdmag.org/download/simple-quorum-drive-freebsd-ctl-ha-beast-storage-system/) OPNsense 17.1.4 released (https://forum.opnsense.org/index.php?topic=4898.msg19359) *** Feedback/Questions gozes asks via twitter about how get involved in FreeBSD (https://twitter.com/gozes/status/846779901738991620) ***

interview guide green writing murder batman plan built wikipedia developers pi programming cows desperate improved initial storage how to inferno began sd open source tutorials gpt linux aspiring numerous programmers raspberry pi editions o2 dragonfly ssl abstraction vax packet unix smp digital ocean bell labs pdp bsd distillation efi freebsd openssl 1mb v10 pull requests openssh alcatel lucent zfs openbsd v9 broadwell nvme ssd raspbian pine64 skylake emmc netbsd lxde allan jude 16mb ixsystems beaglebone black tarsnap dec pdp 128kb 31mb banana pi matt ahrens bsd now dragonflybsd

LCC 165 - Et toi tu scales comment tes données ?

Les Cast Codeurs Podcast

Play Episode Listen Later Mar 15, 2017 108:26

Audrey, Antonio, Emmanuel et Guillaume discutent Google Cloud Next, quelques nouveautés de JDK 9, Docker EE (?!), Cloudbleed, SHAttered, Uber et sa culture poison et comment scaler une architecture horizontalement. Entre autre. Enregistré le 14 mars 2017 Téléchargement de l’épisode LesCastCodeurs-Episode–165.mp3 News Langages Emmanuel le nouveau Java champion !!! 55 nouvelles fonctionalites de JDK 9 jlink, multi jar file, repl, collection factory methods, HTML5 javadoc, SHA–3, G1, semantic versioning etc Construire des JARs multi-release avec Maven Nouvelle version de Groovy 2.4.9 Introduction à CompletableStage en Java Retrofit 2.2 Migration a Swift 3 - cest chaud reflexions sur la backward compatibility de Java Unicode expliqué en 15 minutes Middleware Les librairies Java inratables en 2017 Blockchain Etherium en Java Interview sur l’ORM Doctrine de PHP Une overview de Spanner, la base qui taquine CAP CockroachDB Java EE 8 les dates affinees gRPC donné à la Cloud Native Computing Foundation Lagom 1.3 est sorti Kubernetes et son abstraction du runtime de container WePay et le change data capture Vert.x 3.4.0 Infrastructure Docker EE Cloud Post-mortem d’Amazon S3 Comment AWS voit sa competition Google Cloud Next 2017 Les 100 annonces de Cloud Next Free trial / Free tier amélioré Compute: App Engine Flex (GA), Cloud Functions (beta) et Firebase Functions, new regions, committed use discount, Skylake et 64 vCPU BigData: Dataprep, data transfer service pour BigQuery, Datalab (GA) Databases: Spanner, PostgreSQL Machine Learning: Cloud Machine Learning Engine (GA), video intelligence API, rachat de Kaggle Security: KMS (GA), 2FA, Data Loss Prevention API, Identity-Aware Proxy, Titan security chip Formations Google Cloud sur Coursera Outillage Adopte un desktop Linux par PAG Chrome les dix ans et la genèse du projet Apache Maven 3.5 avec de la couleur ! Gradle 3.4 dépote avec la compilation incrémentale Sécurité Le coût des Ransomware CloudBleed - CloudFlare et l’overrun à un million de dollars Le post-mortem de CloudFlare SHA–1 et la premiere collision: Shattered - les details des chercheurs SHA1 et Linux Google pourrait reporter la publication du code Loi et société et organisation GitHub termes de service Uber et segregation des femmes developpeurs Le premier temoignage Dernières évolutions 1/2 Dernières évolutions 2/2 Antoine Sabot-Durand est star spec lead La transformation ING en equipes microservices 12 startups souhaitent inventer la ville de demain avec la Mairie de Paris et NUMA Tim Berners-Lee: I invented the web. Here are three things we need to change to save it Question crowdcasting Morgan Durand nous pose une question sur la scalabilité horizontale et les données. Conférences Devoxx France les 5–7 avril 2017 Devoxx4Kids Paris le 8 avril 2017 Mix-IT les 20–21 avril 2017 Breizhcamp les 19–21 avril 2017 RivieraDev les 11–12 mai 2017 Web2day 7–9 juin, le CfP est ouvert DevFest Lille 9 juin - inscriptions et CfP ouvert Voxxed Days au Luxembourg le 22 juin Jenkins User Conference Paris - 11 juillet Nous contacter Faire un crowdcast ou une crowdquestion Contactez-nous via twitter https://twitter.com/lescastcodeurs sur le groupe Google https://groups.google.com/group/lescastcodeurs ou sur le site web https://lescastcodeurs.com/ Flattr-ez nous (dons) sur https://lescastcodeurs.com/ En savoir plus sur le sponsoring? sponsors@lescastcodeurs.com

google uber migration swift faire api cfp conf linux java github guillaume scales ing luxembourg shattered derni sha groovy vert donn jars construire kubernetes loi g1 enregistr 2fa html5 mairie spanner mixit grpc bigquery google cloud next gradle jdk flattr wepay skylake cloudbleed sha1 devoxx france

CultCast #256 - Why we’re canceling our MacBook Pro order

Play Episode Listen Later Nov 4, 2016 61:23

This week: we’ve canceled our MacBook Pro order! It’s true, and we’ll you why... Plus: we compare new MacBook Pro’s performance to older models and similarly priced machines; Apple call it quits on external displays; and, the end of an era—one of Mac’s most iconic features gets retired. This episode supported by Casper’s American-made mattresses have just the right amount of memory foam and latex, and people everywhere love them. Learn why and get $50 towards any mattress at Casper.com/cultcast. CultCloth will keep your iPhone 7, Apple Watch, Mac and iPad sparkling clean, and for a limited time you can get 20% off your order with code JETBLACK at CultCloth.co. We also want to give Kevin MacLeod at incompetech.com a thanks for the great music you hear on today's show. On the show this week @erfon / @bst3r / @lkahney Professional Mac Users' Complaints List Grows After 'Disappointing' Apple Event http://www.macrumors.com/2016/11/02/professional-mac-users-complaints-grow-apple-event/ It seems like everyone’s pissed! This week’s notes and links Hiked prices really make you think about what you’re getting, especially compared to how much better it is to what you already have (or what’s available on the market). Raw Power differences/Speed increases in new MBP (compared to 2012 model) We specs aren’t everything as Apple optimizes everything. Felix Schwarz on twitter CPU Only ~30% increase in Geekbench Single and MultiCore increases if buying the same speed processor you have no. Skylake processors are already a year old, but Kaby lake isn't due out until January and Cannonlake due in 2017 as well. GPU You get 100-180% increase in GPU performance over 2012 dedicated GPUs AMD Pro 400 series is used for low power consumption. 4 year old architecture. it’s a mobile version of a budget graphics card AMD claims "the thinnest graphics processor possible” ideal for artists, designers, photographers, filmmakers, visualizers Not on par with Nvidia GPUs Nvidia’s budget card, GTX 1060, performs at 4.2 teraflops a second vs AMD Pro 460’s 1.8 Teraflops. 455’s 1.2 Tflops 450’s 1 Tflop. Nickel and dimed New cables or dongles for all your USB devices, your iPhone, your peripherals and Thunderbolt 2 devices, maybe your existing monitor. No power brick extension cable. Funny enough, Phil Schiller noted the new MacBook Pro still has a 3.5mm headphone jack because it is a "pro machine," but lacks an SD card reader because it's a "cumbersome" slot best left to adapters or wireless transfers. Take into account the $400-$500 price increases, it’s hard to get on board. Next year’s upgrade will address a number of my concerns Much newer Intel CPUs. More availability of USB C and Thunderbolt 3 tech. Lower prices. Touch Bar bugs worked out (if any). Cheaper SSD upgrades. Again, specs aren’t everything. Apple says the new MacBook Pro has: 130% better 3D graphics performance. 60% better gaming performance 57% better video editing performance Performance comparison of the 2012 and 2016 15” Retina MacBook Pro https://pbs.twimg.com/media/Cv-JmoDWYAAlMp7.jpg:large MacBook Pro’s two biggest problems may get fixed in 2017 http://www.cultofmac.com/452087/macbook-pros-two-biggest-problems-may-get-fixed-2017/ The KGI Securities analyst told investors in a recent note that he expects Apple will bring big price-cuts to the MacBook Pro along with some internal upgrades. Kuo also claims the 2017 update will finally give the MacBook Pro up to 32GB of RAM. KabyLake CPUs will be used if Intel’s Cannonlake chips aren’t ready. That will keep max RAM support at 16GB for another year. Price cuts on the next MacBook Pro won’t arrive until the second half of 2017 though. Apple is officially done making displays http://www.cultofmac.com/451644/apple-officially-done-making-standalone-displays/ Buster The days of Apple making its own stand-alone displays for the Mac Mini and Mac Pro are dead. Apple revealed a new 5K 27-inch Thunderbolt 3 display during its “Hello Again” keynote yesterday, only instead of being made by Apple, the company partnered with LG to create the monito The Verge editor-in-chief Nilay Patel tweeted that, after asking about it at the Hello Again event, was informed by Apple that “it’s out of the stand-alone display biz.” Apple did add some tech to LG’s display so that it integrates better with Macs. You can adjust the brightness settings on the LG 5K UltraFine display from your Mac, rather than pushing buttons on the display itself. $1300, or $700 for a 4K model 2016 MacBook Pro loses the iconic startup chime http://www.cultofmac.com/451891/unlikely-origins-macs-startup-chime/ The new MacBook Pro jettisons the iconic F-sharp sound Apple uses to show a Mac is booting up. The use of an arpeggiated chord when you started your Mac dates back to the Macintosh II, when software engineer Mark Lentczner incorporated it into the system. The sound was later revised by Jim Reekes, Apple’s senior software engineer in charge of the audio and system sounds, during the 1990s. Over the years, its tone has changed further and the instrumentation has also varied, The reason for getting rid of the sound instead has to do with the fact that the new laptops now turn on from a fully switched-off mode if they’re opened — meaning you can save all that energy you would otherwise have expended pressing the power button. There’s hope! You can reactivate the sound via a terminal command. Nvidia GTX 1060 Laptop vs Desktop Benchmarks – Is Pascal Really a Game Changer? http://digiworthy.com/2016/09/26/nvidia-gtx-1060-laptop-vs-desktop-benchmarks/ How to restore Mac startup chimes on 2016 MacBook Pro http://www.cultofmac.com/452075/restore-macs-startup-chimes-2016-macbook-pro/ How Faster are the Intel processor and Radeon Pro 450 / 455 / 460 GPU on the New MacBook Pro 15 2016? https://www.techwalls.com/new-macbook-pro-15-processor-radeon-pro-gpu/ Apple Has Received More Online Orders for New MacBook Pro Than Any Previous Generation http://www.macrumors.com/2016/11/02/phil-schiller-new-macbook-pro-interview/

american apple performance price iphone funny 3d speed mac ipads lower game changers intel ram kevin macleod laptops apple watches sd verge casper 4k 5k lg amd macbook pro nickel macs gpu canceling usb c apple event thunderbolt mac pro mac mini kuo touch bar jet black 16gb raw power mbp new macbook pros intel cpus hiked 32gb nilay patel teraflops nvidia gtx skylake multicore tflops retina macbook pro kaby cultcast radeon pro kgi securities

Sac de nodes & hotchips : I.A. quelqu’un ?

Play Episode Listen Later Sep 22, 2016 68:54

Actualités C’était chaud : HotChips 28 A méditer, AMD veut s’imposer par le Zen. NVIDIA : Des nouvelles de Parker, une cousine de la puce automobile PX2 permettra-t-elle à Nintendo de rouler sur ses concurrents ? IBM montre les muscles face à Intel avec ses Power9. Et ARM aussi, avec son ARM V8-A. Et Intel ? Il détaille... Skylake. Mais il soutient la recherche sur la communication entre coeurs. Et il rachète Movidius, un designer de puces (les Myriad) pour l’IA. I.A. Quelqu’un ? Google utilise d’ailleurs ses "TPU" depuis des mois. IBM crée un nouveau type de composant se comportant comme un neurone artificiel. Et HP a aussi sa méthode… Ca pulse autour de la PCM et des memristors ! Sac de nodes Le "Process shrink" apporte ses bienfait partout, même en entré de gamme (NOTE: Les footnotes c’est le pied !), par exemple : le Kirin 650 à 16nm FF+. Alors 14nm ou 16nm FF+ ? Qu’est-ce qu’un "node" ? Crónica de una muerte anunciada : c’est officiel : fin de la miniaturisation après 2021. Keep Calm and Carry On : pendant ce temps, Intel démarre le 10nm ! TSMC prépare le sien pour 2017 et même le 7nm pour 2018 ! Qui ne sera sans doute pas le même que celui d’Intel prévu pour … 2022 ! Participants La chronique des composants est préparée et développée par Guillaume Poggiaspalla Présenté par Guillaume Vendé

google nintendo ibm intel ia zen alors amd keep calm ff sac carry on tsmc nodes quelqu myriad kirin pcm tpu skylake guillaume vend movidius

ThinkComputers Podcast #72 – Thermaltake SPM Platform, New Titan X, Skylake-X, and More!

safe battlefield nvidia gtx battlefield 1 skylake geforce gtx

Play Episode Listen Later Jul 22, 2016 56:20

In this episode of our weekly tech podcast we talk about Thermaltake's Smart Power Management Platform, NVIDIA's announcement of the new Titan X graphics card, new leaks on Intel's Skylake-X and Kaby Lake-X processors and more! The post ThinkComputers Podcast #72 – Thermaltake SPM Platform, New Titan X, Skylake-X, and More! (https://thinkcomputers.org/thinkcomputers-podcast-72-thermaltake-spm-platform-new-titan-x-skylake-x-and-more/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

platform nvidia skylake titan x thermaltake

Episode 5: Kaby Lake-X, Skylake-X, Budget GPU catfight and the world's largest hard drive

The Full Nerd

Play Episode Listen Later Jul 20, 2016 67:10

Brad gives us the full skinny on why the RX 480 makes more sense than the GTX 1060 for most, the crew dives into the rumors of Kaby Lake-X and Skylake-X and whether Intel would eliminate low-end sockets. For show-and-tell in the Builder's Corner, Gordon busts out the world's largest consumer hard drive. Plus audience questions.

budget intel largest builder corner rx hard drive catfight gtx skylake kaby lake

Episode 0: GeForce GTX 1080, Battlefield 1!? and is it safe to build a PC?

The Full Nerd

Play Episode Listen Later May 11, 2016 71:30

This week on the Full Nerd: Gordon Mah Ung, Hayden Dingman and Brad Chacos discuss the glory of Nvidia's new GeForce GTX 1080 and GTX 1070; whether to go Skylake or Broadwell-E and what the what? Battlefield 1? We also answer your burning question.

Avsnitt 65 – ”Lite våld sitter aldrig fel!”

Nördliv - En podcast om spel och nörderi

Play Episode Listen Later Mar 28, 2016 97:54

Medverkande denna gång är: Fredrik, Danny, Lotta, Calle och Robb. I vårt sextiofemte avsnitt så tas följande upp! SPEL I FOKUS Denna gång snackar vi om det gamla NES spelet The Guardian Legend, lite om Elite Dangerous, SNES klassikern Terranigma, There´s Poop in my Soup och Shower with your Dad simulator 2015. - NY SEKTION! - UTMANINGEN! Denna vecka utmanar Lotta någon! UTMANINGSUPPFÖLJNINGEN! Vi följer upp spelet Murdered: Soul Suspect som Lotta fick på sig att spela! Spelnyheter som diskuteras om; Här tar vi upp ang. Microsofts ändring om förlängt stöd för Skylake processorer i Windows 7/8.1, hur Playstation VR sålde slut på 25 sekunder och Face2Face teknologin som ser himla kul ut! Veckans diskussion: Varför gillar gamers gore och våld? Övriga nördämnen Denna vecka snackar vi kring Daredevils andra säsong. Och avslutningsvis går vi igenom och besvarar mail som inkommit. B.la svarar vi på en lyssnares fråga om vilka rollspelskaraktärer vi skulle vara om vi fick nöjet att ge oss ut på äventyr! info@nordlivpodcast.se★ Support this podcast on Patreon ★

ASUS Straight Edge 003 - Q&A session with Intel's Dan Snyder and Aaron Mcgavock

ASUS Straight Edge Podcast

Play Episode Listen Later Mar 16, 2016 49:06

The ASUS Straight Edge podcast returns with a new episode featuring longtime Intel PR manager Dan Snyder, and strategic marketing lead for enthusiasts, Aaron Mcgavock. With 40 years of experience at Intel between them, Dan and Aaron have lots of insights related to tech and media to share. We present our questions and yours, to dig deep and get the nitty gritty details. This is not an episode to miss!

intel cpu dan snyder asus straight edge xeon chipset skylake mcgavock

Chronique des composants : Mobile World Congress 2016

Play Episode Listen Later Feb 28, 2016 37:41

Dans cet épisode, nous revenons sur les annonces du Mobile World Congress 2016 sous un angle "Chronique des composants" c'est-à-dire, en détaillant ce qui fait fonctionner ces appareils. Mais évidemment, on ne parle pas que de ça. On y parle même d'overclocking avec refroidissement par azote liquide. Les sujets Le premier test de la beta de Vulkan sur Talos Principle Les records d'overclocking, c'est toujours rigolo L'overclocking avec azote liquide de Cédric Skylake a d'ailleurs dépassé les 7Ghz ! Qualcomm se lance dans les puces pour IoT Eyeriss, le futur cerveau de votre mobile Les nouveaux CPU mobiles stars du MWC, l'Exynos 8890 du S7 Et le snapdragon 820 du LG G5 ! Et le Kirin 950 du Huawei P9 ! Et le Helio X20 du ... heuuu du Meizu MX6 ? Y'a BASTON ! Nouveaux coeurs pour les microcontroleurs : les Cortex R8 Et les A32 (et 35), qui remplace les A5 et A7 Une sauvegarde éternelle ? Les participants Guillaume Poggiaspalla Cédric Tamboise (@cedsib) Présenté par Guillaume Vendé (@GuillaumeVende)

pr dans qualcomm cpu nouveaux chronique mwc mobile world congress vulkan a5 kirin baston exynos skylake lg g5 a32 huawei p9 7ghz

ThinkComputers Podcast #51 – NZXT Manta, Non-K Skylake Overclocking Blocked, The Division & More!

intel blocked manta overclocking nzxt skylake

Play Episode Listen Later Feb 13, 2016 62:02

In this episode of our weekly tech podcast we talk about our review of NZXT's first small form factor case the Manta, Intel blocking overclocking on non-K Skylake processors, The Division open beta coming up and more! The post ThinkComputers Podcast #51 – NZXT Manta, Non-K Skylake Overclocking Blocked, The Division & More! (https://thinkcomputers.org/thinkcomputers-podcast-51-nzxt-manta-non-k-skylake-overclocking-blocked-the-division-more/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

#21 Manzanas powered by Microsoft

Play Episode Listen Later Feb 13, 2016 104:54

Hoy nos ha salido un podcast más relajado de lo habitual, así que sentaos con tranquilidad para escucharnos hablar de la nueva campaña de Apple para penetrar en la empresa y de lo que realmente va a suponer la familia de procesadores Skylake de Intel para la gama de portátiles de Apple. El tema principal del día es Mail para OS X, sus carencias y cómo podemos solventarlas con algunos plugins.A continuación pasamos a dar un repaso a Default Folder X, una poderosa herramienta que sirve para mejorar la ventana de diálogo para abrir y guardar archivos en las aplicaciones de OS X, pero que es realmente MUCHO más, sobre todo en su nueva versión. Esto nos lleva a un comentario sobre productividad personal, dejando la promesa de abundar en ello en futuros episodios. Para terminar, David muestra su arrepentimiento por mofarse del teclado Logitech de Emilcar y nos trae su propio teclado fetiche, un Kanex.Busca los enlaces de este episodio en http://emilcar.fm, donde también esperamos tus comentarios.

apple microsoft mail esto intel powered mucho busca productividad logitech os x accesorios manzanas emilcar skylake default folder x

#21 Manzanas powered by Microsoft

PC Perspective Podcast Video

Play Episode Listen Later Feb 12, 2016 104:54

apple microsoft mail esto intel powered mucho busca productividad logitech os x accesorios manzanas emilcar skylake default folder x

PC Perspective Podcast 385 - 02/04/16

Play Episode Listen Later Feb 4, 2016 76:38

PC Perspective Podcast #385 - 02/04/2016 Join us this week as we discuss Rise of the Tomb Raider performance, a triple RAID-0 NVMe array and more! You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE. The URL for the podcast is: http://pcper.com/podcast - Share with your friends! iTunes - Subscribe to the podcast directly through the iTunes Store (audio only) Video version on iTunes RSS - Subscribe through your regular RSS reader (audio only) Video version RSS feed MP3 - Direct download link to the MP3 file Hosts: Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano Program length: 1:16:38 Join our spam list to get notified when we go live! We’re on Patreon! Week in Review: 0:04:10 Rise of the Tomb Raider: AMD and NVIDIA Performance Results 0:11:51 Triple M.2 Samsung 950 Pro Z170 PCIe NVMe RAID Tested - Why So Snappy? 0:33:45 AMD Refreshes Q1 2016 Offerings: New APUs, CPUs, and Coolers 0:42:17 "What's the Point of Steam OS?" 0:44:25 Winner: EVGA Winter 2016 Prize Pack and Giveaway News items of interest: 0:46:35 Gigabyte adds full GIMPS and Prime95 compatibility to Skylake processors 0:48:40 So That's Where Jim Keller Went To... Tesla Motors… 0:54:40 AMD FirePro S-Series Introduces Hardware-Based GPU Virtualization 0:56:15 Who's a pretty boy? Is it you Fallout? 0:58:40 OCZ Launches Trion 150, Successor to Trion 100 SATA SSD, Using 15nm Flash Hardware/Software Picks of the Week: 1:01:30 Ryan: Peavey USB Audio Interface 1:04:00 Jeremy: This, a VR headset, a spaceship and you 1:07:32 Josh: Use these every day. Best 2.1 Evar. 1:09:40 Allyn: I’m gonna let you finish, but this is the best PC puzzle game of all time. (The Witness) http://pcper.com/podcast http://twitter.com/ryanshrout and http://twitter.com/pcper Closing/outro Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!

video witness pc vr fallout samsung raid tomb raider week in review successors cpus triple m tesla motors gigabytes trion steamos nvme evar gimps prize pack skylake sata ssd pc perspective podcast

The Sams Report EP 25 Skylake Woes, Xbox And Redstone

The Sams Report

Play Episode Listen Later Jan 22, 2016 27:36

On this episode, I cover the all the news that Microsoft announced this week including limited support for Skylake CPUs, Xbox news, a new build of Redstone (insider info) and a lot more.

microsoft xbox woes sams redstone skylake

125: DevSummits, Core and the Baldwin

new year structure damage atari internet explorer postgres openssl skylake

Play Episode Listen Later Jan 20, 2016 133:49

This week on the show, we will be talking to FreeBSD developer and former core-team member John Baldwin about a variety of topics, including running a DevSummit, everything you needed or wanted to know. Coming up right now on BSDNow, the place to B...SD. This episode was brought to you by Headlines FreeBSD server retired after almost 19 years (http://www.theregister.co.uk/2016/01/14/server_retired_after_18_years_and_ten_months_beat_that_readers/) We've heard stories about this kind of thing before, that box that often sits under-appreciated, but refuses to die. Well the UK register has picked up on a story of a FreeBSD server finally being retired after almost 19 years of dedicated service. “In its day, it was a reasonable machine - 200MHz Pentium, 32MB RAM, 4GB SCSI-2 drive,” Ross writes. “And up until recently, it was doing its job fine.” Of late, however the “hard drive finally started throwing errors, it was time to retire it before it gave up the ghost!” The drive's a Seagate, for those of you looking to avoid drives that can't deliver more than 19 years of error-free operations. This system in particular had been running FreeBSD 2.2.1 over the years. Why not upgrade you ask? Ross has an answer for that: “It was heavily firewalled and only very specific services were visible to anyone, and most only visible to our directly connected customers,” Ross told Vulture South. “By the time it was probably due for a review, things had moved so far that all the original code was so tightly bound to the operating system itself, that later versions of the OS would have (and ultimately, did) require substantial rework. While it was running and not showing any signs of stress, it was simply expedient to leave sleeping dogs lie.” All in all, an amazing story of the longevity of a system and its operating system. Do you have a server with a similar or even greater uptime? Let us know so we can try and top this story. *** Roundup of all the BSDs (https://www.linuxvoice.com/group-test-bsd-distros/) The magazine LinuxVoice recently did a group test of a variety of “BSD Distros”. Included in their review were Free/Open/Net/Dragon/Ghost/PC It starts with a pretty good overview of BSD in general, its starts and the various projects / forks that spawned from it, such as FreeNAS / Junos / Playstation / PFSense / etc The review starts with a look at OpenBSD, and the consensus reached is that it is good, but does require a bit more manual work to run as a desktop. (Most of the review focuses on desktop usage). It ends up with a solid ⅘ stars though. Next it moves into GhostBSD, discusses it being a “Live” distro, which can optionally be installed to disk. It loses a few points for lacking a graphical package management utility, and some bugs during the installation, but still earns a respectable ⅗ stars. Dragonfly gets the next spin and gets praise for its very-up to date video driver support and availability of the HAMMER filesystem. It also lands at ⅗ stars, partly due to the reviewer having to use the command-line for management. (Notice a trend here?) NetBSD is up next, and gets special mention for being one of the only “distros” that doesn't do frequent releases. However that doesn't mean you can't have updated packages, since the review mentions pkgsrc and pkg as both available to customize your desktop. The reviewer was slightly haunted by having to edit files in /etc by hand to do wireless, but still gives NetBSD a ⅗ overall. Last up are FreeBSD and PC-BSD, which get a different sort of head-to-head review. FreeBSD goes first, with mention that the text-install is fairly straight-forward and most configuration will require being done by hand. However the reviewer must be getting use to the command-line at this point, because he mentions: “This might sound cumbersome, but is actually pretty straightforward and at the end produces a finely tuned aerodynamic system that does exactly what you want it to do and nothing else.” He does mention that FreeBSD is the ultimate DIY system, even to the point of not having the package management tools provided out of box. PC-BSD ultimately gets a lot of love in this review, again with it being focused on desktop usage this follows. Particularly popular are all the various tools written to make PC-BSD easier to use, such as Life-Preserver, Warden, the graphical installer and more. (slight mistake though, Life-Preserver does not use rsync to backup to FreeNAS, it does ZFS replication) In the end he rates FreeBSD ⅘ and PC-BSD a whopping 5/5 for this roundup. While reviews may be subjective to the particular use-case being evaluated for, it is still nice to see BSD getting some press and more interest from the Linux community in general. *** OpenBSD Laptops (http://www.tedunangst.com/flak/post/openbsd-laptops) Our buddy Ted Unangst has posted a nice “planning ahead” guide for those thinking of new laptops for 2016 and the upcoming OpenBSD 5.9 He starts by giving us a status update on several of the key driver components that will be in 5.9 release“5.9 will be the first release to support the graphics on Broadwell CPUs. This is anything that looks like i5-5xxx. There are a few minor quirks, but generally it works well. There's no support for the new Skylake models, however. They'll probably work with the VESA driver but minus suspend/resume/acceleration (just as 5.8 did with Broadwell).” He then goes on to mention that the IWM driver works well with most of the revisions (7260, 7265, and 3160) that ship with broadwell based laptops, however the newer skylake series ships with the 8260, which is NOT yet supported. He then goes on to list some of the more common makes and models to look for, starting with the broadwell based X1 carbons which work really well (Kris gives +++), but make sure its not the newer skylake model just yet. The macbook gets a mention, but probably should be avoided due to broadcom wifi The Dell XPS he mentions as a good choice for a powerful (portable) desktops *** Significant changes from NetBSD 7.0 to 8.0 (https://www.netbsd.org/changes/changes-8.0.html) Updated to GCC 4.8.5 Imported dhcpcd and replaced rtsol and rtsold gpt(8) utility gained the ability to resize partitions and disks, as well as change the type of a partition OpenSSH 7.1 and OpenSSL 1.0.1q FTP client got support for SNI for https Imported dtrace from FreeBSD Add syscall support Add lockstat support *** Interview - John Baldwin - jhb@freebsd.org (mailto:jhb@freebsd.org) / @BSDHokie (https://twitter.com/BSDHokie) FreeBSD Kernel Debugging News Roundup Dragonfly Mail Agent spreads to FreeBSD and NetBSD (https://www.dragonflydigest.com/2016/01/18/17508.html) DMA, the Dragonfly Mail Agent is now available not only in Dragonfly's dports, but also FreeBSD ports, and NetBSD pkgsrc “dma is a small Mail Transport Agent (MTA), designed for home and office use. It accepts mails from locally installed Mail User Agents (MUA) and delivers the mails either locally or to a remote destination. Remote delivery includes several features like TLS/SSL support and SMTP authentication. dma is not intended as a replacement for real, big MTAs like sendmail(8) or postfix(1). Consequently, dma does not listen on port 25 for incoming connections.” There was a project looking at importing DMA into the FreeBSD base system to replace sendmail, I wonder of the port signals that some of the blockers have been fixed *** ZFS UEFI Support has landed! (https://svnweb.freebsd.org/base?view=revision&revision=294068) Originally started by Eric McCorkle Picked up by Steven Hartland Including modularizing the existing UFS boot code, and adding ZFS boot code General improvements to the EFI loader including using more of libstand instead of containing its own implementations of many common functions Thanks to work by Toomas Soome, there is now a Beastie Menu as part of the EFI loader, similar to the regular loader As soon as this was committed, I added a few lines to it to connect the ZFS BE Menu to it, thanks to all of the above, without whom my work wouldn't be usable It should be relatively easy to hook my GELI boot stuff in as a module, and possibly just stack the UFS and ZFS modules on top of it I might try to redesign the non-EFI boot code to use a similar design instead of what I have now *** How three BSD OSes compare to ten Linux Distros (http://www.phoronix.com/scan.php?page=article&item=3bsd-10linux) After benchmarking 10 of the latest Linux distros, Phoronix took to benchmarking 3 of the big BSDs DragonFlyBSD 4.4.1 - The latest DragonFly release with GCC 5.2.1 and the HAMMER file-system. OpenBSD 5.8 - OpenBSD 5.8 with GCC 4.2.1 as the default compiler and FFS file-system. PC-BSD 10.2 - Derived off FreeBSD 10.2, the defaults were the Clang 3.4.1 compiler and ZFS file-system. In the SQLite test, PCBSD+ZFS won out over all of the Linux distros, including those that were also using ZFS In the first compile benchmark, PCBSD came second only to Intel's Linux distro, Clear Linux. OpenBSD can last, although it is not clear if the benchmark was just comparing the system compiler, which would be unfair to OpenBSD In Disk transaction performance, against ZFS won the day, with PCBSD edging out the Linux distros. OpenBSD's older ffs was hurt by the lack of soft updates, and DragonFly's Hammer did not perform well. Although in an fsync() heavy test, safety is more important that speed As with all benchmarks, these obviously need to be taken with a grain of salt In some of them you can clearly see that the ‘winner' has a much higher standard error, suggesting that the numbers are quite variable *** OPNSense 15.7.24 Released (https://opnsense.org/opnsense-15-7-24-released/) We are just barely into the new year and OPNSense has dropped a new release on us to play with. This new version, 15.7.24 brings a bunch of notable changes, which includes improvements to the firewall UI and a plugin management section of the firmware page. Additionally better signature verification using PKG's internal verification mechanisms was added for kernel and world updates. The announcement contains the full rundown of changes, including the suricata, openvpn and ntp got package bumps as well. *** Beastie Bits A FreeBSD 10 Desktop How-to (https://cooltrainer.org/a-freebsd-desktop-howto/) (A bit old, but still one of the most complete walkthroughs of a desktop FreeBSD setup from scratch) BSD and Scale 14 (http://fossforce.com/2016/01/bsd-ready-scale-14x/) Xen support enabled in OpenBSD -current (http://undeadly.org/cgi?action=article&sid=20160114113445&mode=expanded) Feedback/Questions Matt - Zil Sizes (http://slexy.org/view/s20a0mLaAv) Drin - IPSEC (http://slexy.org/view/s21qpiTF8h) John - ZFS + UEFI (http://slexy.org/view/s2HCq0r0aD) Jake - ZFS Cluster SAN (http://slexy.org/view/s2VORfyqlS) Phillip - Media Server (http://slexy.org/view/s20ycRhUkM) ***

live uk interview guide built diy os scale released remote hammer developers intel significant storage how to sd open source baldwin tutorials ui linux roundup warden dragonfly dma derived gcc geli digital ocean ffs x1 seagate bsd sqlite vesa efi xen freebsd openssl smtp clang sni iwm ufs zfs openbsd broadwell skylake freenas pkg john baldwin mtas netbsd allan jude phoronix clear linux bsds ixsystems kris moore tarsnap ghostbsd pc bsd dragonflybsd

Episode 249: Internet of Threats | TechSNAP 249

TechSNAP

Play Episode Listen Later Jan 14, 2016 90:58

A Critical OpenSSH flaw can expose your private keys, a new WiFi spec for IoT devices, that has all the classic issues & Intel’s SkyLake bug. Plus your feedback, our answers, a rockin’ round up & so much more!

internet threats wifi intel iot skylake techsnap

Internet of Threats | TechSNAP 249

TechSNAP Large Video

Play Episode Listen Later Jan 14, 2016 90:58

stories internet technology news debate microsoft threats windows wifi intel analysis jupiter iot broadcasting linux apache encryption ssl bsd skylake techsnap

Internet of Threats | TechSNAP 249

TechSNAP Mobile Video

Play Episode Listen Later Jan 14, 2016 90:58

stories internet technology news debate microsoft threats windows wifi intel analysis jupiter iot broadcasting linux apache encryption ssl bsd skylake techsnap

Control Structure #99: Critical CPU Damage

Control Structure

Play Episode Listen Later Jan 13, 2016 63:00

Andrew and Steve talk briefly talk about what happened over New Year's, like watching transmissions, Skylake, Internet Explorer end of support, OpenSSL funding, Postgres, reponsive design, and Atari.

KBMOD Podcast – Episode 218

The KBMOD Podcast

Play Episode Listen Later Dec 15, 2015 110:11

This week's podcast might be shy one APL, but his absence did not stop the rest of the team from providing you with a bottomless meme-soaked rabbit hole of PC gaming news to tumble down. Tribes Ascend received its first update in more than two years, and it is a substantial one. If your Skylake [&hellip

pc apl skylake tribes ascend

NZ Tech Podcast 254: Killer USB sticks, Lightning Lab Manufacturing, Skylake, The Best Smartphone cameras

NZ Tech Podcast

Play Episode Listen Later Oct 20, 2015 64:36

This episode - Tesla's free autonomous driving update, the best camera phones, Intel Skylake, Sony Xperia Z5 and Z5 Compact, Tom Tom Spark fitness watch, killer USB sticks, Blackberry Priv Android handset and hear all about Lightning Lab Manufacturing. Running time : 1:04:35

running killers tesla smartphones lightning manufacturing usb cameras tech podcast usb sticks skylake intel skylake sony xperia z5

Chronique des composants (octobre 2015)

bonne chronique tipeee skylake

Play Episode Listen Later Oct 14, 2015 64:42

La tant attendue Chronique des composants est de retour ! On parle réalité virtuelle, de coeurs, d'exaflops, de Skylake, de mémoire et même d'énergie gratuite (ou presque) ! Bonne écoute ! Pour celles et ceux qui ne l'auraient pas encore découvert, nous avons lancé un financement participatif pour nous aider à faire vivre le podcast. Pour en savoir plus, rendez-vous sur la page Tipeee de Tech Café. A noter que nous ne prélevons que les épisodes dits "classiques" et pas les "hors-série" ou les épisodes de la "Chronique des composants". Si nos émissions vous plaisent, peut-être devriez-vous jeter un coup d'oeil à ce financement, accessible à partir de l'équivalent d'un café par mois.

#12 Sólo 4K

apple os mac bienvenidos parece busca webcams aprenderemos el capit cleanmymac logitech c920 skylake

Play Episode Listen Later Oct 8, 2015 28:13

Bienvenidos a nuestra cita quincenal sobre el Mac y sólo el Mac. Parece que Apple nos traerá por fin novedades de hardware para la próxima semana, aunque no tengamos todavía procesadores Skylake. Os comento también mis experiencias con la webcam Logitech C920 y os hablo sobre una app que puede mejorar mucho esta y todas las webcams. A continuación paso a exponeros mis temores ante la falta de actualizaciones para OS X El Capitán de las apps propias de Apple. Aprenderemos también cómo CleanMyMac3 puede (o no) liberar espacio de nuestro iTunes. Y para terminar hablaremos de tres pequeñas novedades de El Capitán que no deberían pasarnos desapercibidas.Busca los enlaces de este episodio en http://emilcar.fm, donde también espero tus comentarios.

#12 Sólo 4K

apple os mac bienvenidos parece busca webcams aprenderemos el capit cleanmymac logitech c920 skylake

Play Episode Listen Later Oct 8, 2015 28:13

Podcast #060

Ni Tan Fanboys Podcast

Play Episode Listen Later Sep 16, 2015 66:14

En este nuevo podcast en solitario sin Monky, comentamos con respecto a la Keynote del evento Apple Special Event de Septiembre de 2015 en el cual Apple presentó el nuevo Apple TV, el iPad Pro junto con la actualización del iPad Mini 4 y los iPhone 6S y 6S Plus. En la sección Noticias de Apple, les comento acerca de la alianza de Apple con Cisco, de las nuevas aplicaciones que está desarrollando en conjunto con IBM, de la eliminación en forma remota del canal de YouTube en el Apple TV 2 y de cómo puede ser reinstalado gracias a aTV Flash Black de Firecore. En la sección Internet, les comento respecto de la nueva familia de procesadores Skylake de Intel. En la sección iOS Apps comentamos respecto de la aplicación Setlist. En la sección Mac Apps, les comento respecto de la aplicación CleanMyMac 3. En la sección Gadgets les comento respecto de una carcaza para MacBook Pro comprada en AliExpress. Finalmente en la sección Bookmarks les comento del nuevo blog basado en Tumblr que está escribiendo mi hijo Joaquín Montes cuya temática es la Ciencia Espacial.

Podcast #060

Ni Tan Fanboys Podcast

Play Episode Listen Later Sep 16, 2015 66:14

Intel Skylake - Episode 94

Piltch Point (Video)

Play Episode Listen Later Sep 6, 2015 14:25

This week, Avram Piltch discusses what we didn't get to hear about at this year's Intel Developer Forum: Skylake. Waiting for IDF 2015, the company talked about all of the cool new features of the next generation of processors, including its implementation into the Core M, Pentium and Celeron hardware. He also discusses some of the new computers that are taking advantage of the new Skylake architecture.

idf pentium skylake celeron avram piltch intel skylake core m

The AnandTech Podcast: Episode 35

The AnandTech Podcast - M4A Feed

Play Episode Listen Later Aug 31, 2015 109:25

AnandTech Podcast #35: The summer season has been busier than usual. Microsoft released Windows 10, it's latest operating system, as a free upgrade to current Windows 7 and Windows 8 users, promising it to be the latest and greatest from Microsoft, but also why it might be the last full release of an operating system from them. Brett's indepth review of Windows 10's features is now available on the AnandTech website, with a link below. Hot on the heels of Windows 10 was Intel's launch of its new Skylake processor microarchitecture, which came in two desktop processors only to begin with. This was followed up by a more in-depth discussion of what Skylake actually does at Intel's Developer Forum, and in this podcast we discuss what Skylake brings to the table as well as the performance and our interpretation of how Intel's vision is set to develop in the coming years.

microsoft windows intel anandtech skylake

ThinkComputers Podcast #26 – Gigabyte Z170X-Gaming G1, GTX 950, Skylake, eSport & More

gaming esports esport gigabytes gtx skylake nvidia geforce gtx

Play Episode Listen Later Aug 22, 2015 55:25

This week on our Weekly Tech Podcast we talk about our review of the Gigabyte Z170X-Gaming G1 motherboard, the launch of the NVIDIA GeForce GTX 950, more Skylake platform details, eSports & more! The post ThinkComputers Podcast #26 – Gigabyte Z170X-Gaming G1, GTX 950, Skylake, eSport & More (https://thinkcomputers.org/thinkcomputers-podcast-26-gigabyte-z170x-gaming-g1-gtx-950-skylake-esport-more/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

Skylake将会给下一代Mac带来什么？FMiT.VOL 142

FMiT

Play Episode Listen Later Aug 8, 2015 62:57

skylake

ThinkComputers Podcast #24 – Skylake is Here!

Play Episode Listen Later Aug 7, 2015 68:18

This week on our Weekly Tech Podcast we talk about the arrival of Intel's Skylake platform and our review of the Core i7-6700K processor, ASUS A170-A motherboard, Gigabyte Z170A Gaming M5 motherboard, and much more! The post ThinkComputers Podcast #24 – Skylake is Here! (https://thinkcomputers.org/thinkcomputers-podcast-24-skylake-is-here/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

skylake 6700k

129: Tap-to-Click Wizard

Accidental Tech Podcast

Play Episode Listen Later Aug 6, 2015 137:44

Skylake desktop CPUs launch Casey joins 2012 "Change your apps name. Not that big of a deal." TrackPoint TrackPoint Mouse Leap Motion 2015 iPhone predictions Naked robotic core It's Ramifications! Carrier subsidies and pricing games AT&T Next Post-show: Repairing iPads Hot water heaters Marco's new camera Sony A7R II Sony FE 35mm f/2.8 Sony FE 55mm f/1.8 Sony FE 90mm f/2.8 macro Canon 5D Mark II Nikon D750 Casey's camera setup Sponsored by: Cards Against Humanity: A free party game for horrible people. This week's toaster Harry's: An exceptional shave at a fraction of the price. Use code ATP for $5 off your first purchase. Hover: The best way to buy and manage domain names. Use coupon code INFINITETIMESCALE for 10% off.

change wizard atp carrier skylake

ThinkComputers Podcast #23 – Windows 10, Zotac GTX 980 Ti AMP!, Skylake & More!

windows windows 10 zotac skylake

Play Episode Listen Later Jul 31, 2015 56:09

This week on our Weekly Tech Podcast we talk about our review of the the release of Windows 10, Zotac's GeForce GTX 980 Ti AMP! graphics card, some upcoming Skylake products, and much more! The post ThinkComputers Podcast #23 – Windows 10, Zotac GTX 980 Ti AMP!, Skylake & More! (https://thinkcomputers.org/thinkcomputers-podcast-23-windows-10-zotac-gtx-980-ti-amp-skylake-more/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

ThinkComputers Podcast #22 – Sapphire Tri-X R9 390X, Deepcool, Intel Skylake & More

windows intel redmond windows 10 surface pro rtm skylake

Play Episode Listen Later Jul 24, 2015 73:11

This week on our Weekly Tech Podcast we talk about our review of the Sapphire Tri-X R9 390X graphics card, a few products from Deepcool, Intel Skylake overclocks, upcoming Skylake motherboards, and much more! The post ThinkComputers Podcast #22 – Sapphire Tri-X R9 390X, Deepcool, Intel Skylake & More (https://thinkcomputers.org/thinkcomputers-podcast-22-sapphire-tri-x-r9-390x-deepcool-intel-skylake-more/) appeared first on ThinkComputers.org (https://thinkcomputers.org) .

skylake tri x intel skylake

The Redmond Report - EP 2, Windows 10 RTM and Skylake

The Sams Report

Play Episode Listen Later Jul 17, 2015 44:38

On this episode we talk about Windows and its march towards RTM, Surface Pro 4, and Intel's new Skylake CPUs.

Chronique des composants (juillet 2015)