Cloud Masters is a no-fluff cloud podcast distilling the learnings from 140,000+ cloud challenges and customer experiences into tangible tips for navigating the cloud.
We cover how Istio Ambient Mesh eliminates sidecar proxies to significantly reduce Kubernetes resource consumption. This episode covers the architectural differences between traditional service mesh and ambient mesh, practical migration strategies for different workload types, key metrics for measuring performance improvements, and real-world operational benefits like simplified troubleshooting and easier version upgrades.
We cover why AWS is now requiring MFA on all AWS account root users, and demo the new centralized root credential management feature that makes it easy to manage credentials across member accounts.
Learn how AWS database experts diagnose and solve the most common RDS performance bottlenecks, from storage configuration to instance type selection. Discover practical strategies to optimize your database costs while ensuring reliability, including when to choose between RDS and Aurora, how to pick the right storage type, and which metrics actually matter for monitoring.
AWS Analytics expert Swapnil Bhoite joins us to break down of Amazon S3 Tables. From comparing Parquet and Apache Iceberg formats to explaining critical features like compaction and snapshot management, Swapnil explores how this fully-managed service streamlines data lake operations. Learn when to adopt S3 Tables, understand its cost-performance benefits, and discover key migration considerations from existing Glue catalog implementations — essential knowledge for teams looking to scale their analytics workloads on AWS.
Learn actionable strategies for handling AWS security breaches from detection through prevention. AWS security experts share real incident response experiences, containment tactics, and practical tips for securing your AWS infrastructure.
Learn when Amazon Aurora DSQL is the right choice for your applications, what changes you'll need to make during migration, and how to handle its unique approach to optimistic locking.
We're joined by two AWS MAP experts to delve into the AWS Migration Acceleration Program (MAP). We go through its three phases: Assess, Mobilize, and Migrate & Modernize; and the program's framework, funding opportunities, and common misconceptions. More specifically, we discuss the importance of accurate capacity planning, TCO calculations, and tagging for successful migrations. We also cover the critical role of project management, stakeholder involvement, and adhering to AWS Well-Architected Framework principles. The conversation touches on challenges in modernizing legacy applications, the significance of proper landing zone setup, and strategies for post-migration optimization. Finally, we conclude with best practices for cloud migrations, emphasizing the importance of involving skilled cloud architects and planning for the unpredictable nature of migration projects. Key Moments: 00:00 - Introduction 00:56 - What is the AWS MAP Program? 01:59 - Phases of AWS MAP 07:32 - Qualifying for MAP 09:42 - Importance of the TCO calculation 14:23 - Underestimating the migration effort 18:18 - Don't forget security costs in your TCO 19:31 - Cloud migration strategies 24:53 - AWS MAP tagging 30:02 - Landing zones 32:15 - Migrating outdated on-prem software 35:14 - Caveats with the TCO calculator 38:50 - Factoring in modernization work 40:43 - Unforeseen issues post-migration 43:27 - Disaster recovery planning 44:50 - AWS MAP best practices
In this episode, we dive into Amazon CloudFront, exploring its benefits, use cases, and cost optimization strategies. Specifically, we go into the importance of Average Object Size (AOS) when wanting to sign a CloudFront PPA. We also discuss how using CloudFront saves you on data transfer costs compared to alternative solutions, its versatility in handling both static and dynamic content, and the importance of page-loading time for user experience. Finally, we conclude with an examination of security considerations when using CloudFront, including strategies for mitigating DDoS attacks while keeping costs in check.
We cover the 5 most significant Kubernetes updates of 2024: Gateway API, Pod in-place resizing, Container resource-based autoscaling, Support for Swap, and User Namespaces.
Everything you need to know about Cloud Marketplaces: Best practices for listing and co-selling, pitfalls to avoid, and lessons learned from our guests' own experiences helping companies navigate marketplace listings.
Three Technical Account Managers (TAMs) from DoiT share real-life stories about overspending on cloud and managing cloud commitment drawdowns, with practical lessons and strategies for optimizing your own cloud costs.
We discuss the top security risks to be aware of when implementing LLMs in your product, and how to prevent them from occurring in the first place.
We're joined by two BigQuery specialists to cover how to transition from flat-rate to BigQuery Editions and on-demand pricing without overspending.
Covering when to use Aurora vs. RDS, common challenges when optimizing Aurora after migrating from RDS, I/O optimization and when I/O-optimized makes sense, and how to architect your Aurora databases for cost optimization.
Kubernetes specialist Alfred Tommy joins us to recap Kubecon Europe 2024, including Istio Ambient Mesh's sidecarless model vs. LinkerD, security tradeoffs between the two, K8sGPT, and more!
From initial exploration to integration, we uncover the evolution of objectives, unexpected discoveries, and challenges faced along the way by two startups in implementing Gen AI in production.
ML and AI specialists Eduardo Mota and Sascha Heyer join us to explore the complexities of observability of LLM-powered features. Packed with tons of real-life customer anecdotes and best practices, they discuss the challenges and strategies for monitoring Gen AI systems, emphasizing the importance of metrics in understanding system interactions, especially given Gen AI's non-deterministic nature.
Covering how Karpenter extends Kubernetes' native autoscaling functionality, why it is so useful for Kubernetes users, and where it makes sense to use it.
Exploring the impact that Gen AI will have on data pipelines and data engineering overall.
Covering the purpose and use cases of KEDA, best practices, common pitfalls, and wrapping up with a scale-from-zero KEDA demo.
We cover the ever-growing importance of Large Language Models (LLMs) in applications, how LLM costs can easily compound once in production, and breaking down the costs associated with using LLMs.
We discuss why AWS is now charging for previously-free public IPv4 addresses, and how to analyze and minimize the impact of this change on your AWS bill.
In this episode, we go over how to optimize BigQuery for performance and cost, as well as the other services you would use alongside it for data ingestion, processing, and consumption.