In Season 4, Episode 42, Karl and Jon are joined by AWS Community Builder Sujal Shaw to discuss a range of new AWS developments. Topics include the New AWS Region Planner for easier regional planning and global deployments, EC2 Auto Scaling warm pool support for mixed instance policies, the AWS European Sovereign Cloud white paper, and the new underwater AWS Fastnet cable featuring robust armoring. They also explore the challenges younger developers face when deploying to AWS. The team was delighted to make it through the episode without any gremlins—despite Karl’s makeshift mobile studio setup in Madrid!

04:12 - AWS Region Planner

AWS introduced a new tool for centralized information on service and feature availability across regions. It helps architects avoid feature parity issues between regions and plan global deployments more efficiently. The tool shows availability status (available, planning, not expanding, or planned) for services and features in different regions.

11:34 - EC2 Auto Scaling Warm Pool Support

AWS announced warm pool support for auto scaling groups with mixed instance policies. This feature combines the benefits of warm pools (faster instance launch) with mixed instances (flexibility in instance types). It allows for both speed and cost savings by keeping a mix of pre-warmed instances ready to launch instantly using the most effective combination of on-demand, spot, or savings plan instances.

16:59 - AWS European Sovereign Cloud

AWS published a white paper detailing the upcoming European Sovereign Cloud. It will be operated by EU residents and citizens, with no critical dependencies on non-EU infrastructure. The cloud aims to address data residency and sovereignty concerns for European organizations, particularly government and regulated industries.

25:30 - AWS Fastnet Underwater Cable

Amazon is equipping a new underwater cable (AWS Fastnet) with robust armoring to prevent cuts. The cable will run from Maryland, US to Ireland, capable of transporting over 320 terabytes per second. The armoring aims to protect against potential threats like fishing anchors or intentional damage, improving durability and reducing maintenance downtime.

32:26 - Challenges of Deploying to AWS

Corey Quinn's article discusses the complexity of deploying to AWS, arguing that younger developers may not tolerate the steep learning curve. The article highlights the numerous steps and services (IAM, VPCs, etc.) required before actual development can begin. It suggests that easier deployment options like Vercel or Heroku, which abstract away complexity, may be more appealing to new developers.

In Season 4, Episode 41, Karl and Jon are joined by AWS Community Builder Fabien Zucchet to discuss AWS Lambda’s increased payload size for asynchronous invocations and what it means for developers building event-driven applications. They cover the new metrics dashboard for AWS Step Functions, best practices for handling sensitive log data in Amazon CloudWatch, and insights into a major AWS service outage and its widespread impact. The episode wraps up with a look at Amazon’s stock jump despite recent challenges, as the trio reflect on the resilience of cloud providers in a constantly evolving market. And then the guys joked about how their video call issues were secretly caused by discussing cloud outages, as if the streaming platform was taking revenge.

03:57 - AWS Lambda payload size increase

AWS Lambda increased its maximum payload size from 256KB to 1MB for asynchronous invocations. The speakers discussed potential use cases and pricing implications, noting that while it solves a problem, it may not be widely necessary and could lead to increased costs for some users.

13:21 - AWS Step Functions metrics dashboard

AWS announced a new metrics dashboard for Step Functions. The speakers viewed this as a positive development, improving observability and providing better insights for users, especially those operating at scale.

17:19 - Handling sensitive log data in Amazon Cloudwatch

The article discussed methods for masking sensitive information in Cloudwatch logs. The speakers highlighted the importance of this feature for compliance with regulations like GDPR and protecting personally identifiable information (PII).

23:56 - AWS service outage

A major AWS outage occurred due to a DNS race condition, affecting multiple services. The speakers discussed the complexity of operating at AWS's scale and the cascading effects of the outage on various dependent services.

33:29 - Amazon's stock performance

Despite recent challenges including the service outage and layoffs, Amazon's stock jumped due to accelerated AWS cloud growth. The speakers discussed how layoffs can paradoxically lead to stock increases and the continued strong performance of AWS as a key driver of Amazon's success.

In Season 4, Episode 40, Karl and Jon are joined by AWS Community Builder Jason Wood to discuss the recent AWS outage in the US-East-1 region and its impact across multiple services. They cover simplified model access in Amazon Bedrock, the introduction of the Claude 4.5 Haiku model, the launch of Amazon EC2 Capacity Manager, new automated CloudWatch dashboards for analyzing log usage, and the latest AWS service availability updates, including those entering maintenance or retirement. Eventually, the conversation takes a lighter turn as the guys debate which countries Jon must speak in to officially graduate from a national to international conference speaker.

09:08 - Simplified model access in Amazon Bedrock

AWS has simplified access to non-AWS models in Bedrock, removing the need for manual activation in most cases. This change streamlines the process of using AI models, though some models like Anthropic's still require a first-time usage form.

14:49 - Claude 4.5 Haiku model in Amazon Bedrock

Anthropic has released Claude 4.5 Haiku, a more cost-effective model with performance comparable to the older Sonnet version. The speakers discussed the naming conventions and the importance of upgrading from older models before they reach end-of-life.

18:51 - Amazon EC2 Capacity Manager

This new feature provides a single interface for monitoring, analyzing, and managing EC2 capacity usage. It combines data from various existing tools, offering a comprehensive view of capacity trends, cost optimization opportunities, and usage metrics.

24:15 - Enhanced automatic dashboard for CloudWatch logs

AWS introduced an enhanced automatic dashboard for analyzing CloudWatch log usage. This feature provides detailed insights into log ingestion, API calls, and cost-related metrics, making it easier to optimize log management and associated costs.

30:32 - AWS service availability updates

AWS announced changes to the availability status of several services. 19 services are moving to maintenance mode, and 4 are entering sunset phase. The speakers discussed the implications of these changes and potential alternatives for affected services.

In Season 4, Episode 38, Karl and Jon welcome AWS Community Builder Luis Valdivia. They discuss the new ECS managed instances for containerized applications on AWS, the cost-effectiveness of serverless architecture at scale, and the latest Anthropic Claude Sonnet 4.5 model now available in Amazon Bedrock. They also cover the general availability of the AWS Knowledge MCP server and a recent social engineering attack on a software platform that exploited AWS domain registration. The episode wraps up with a lighthearted moment as the hosts spend ten minutes trying to determine whether Jon’s background had actually changed or not.

06:05 - ECS managed instances for containerized applications

AWS introduced a new way to run containers called ECS managed instances. This option sits between unmanaged EC2 instances and Fargate, offering more control than Fargate but less management overhead than unmanaged EC2. The pricing model is based on instance type, with a management fee added. This new option provides more flexibility but also adds complexity to the decision-making process for container deployment.

16:17 - Serverless cost-effectiveness at scale

An article by AWS hero Evandro Pires argues that serverless is not inherently expensive at scale, but rather becomes costly when implemented incorrectly. The discussion highlights that serverless encompasses more than just Lambda functions and that proper architecture is crucial for cost-effective serverless deployments. The speakers agree that bad architecture, rather than the serverless approach itself, is often the root cause of high costs.

23:44 - Anthropic Claude Sonnet 4.5 model in Amazon Bedrock

AWS announced the availability of Anthropic's latest AI model, Claude Sonnet 4.5, in Amazon Bedrock. This model is described as Anthropic's most intelligent, particularly for coding and complex agents. The speakers discuss the benefits of having this model integrated into AWS's ecosystem, including enhanced security and data privacy. They also note the incremental improvements over previous versions and its capabilities compared to other models.

30:11 - AWS Knowledge MCP server

AWS released the Knowledge MCP server, which allows large language models (LLMs) to access AWS documentation and knowledge bases. This tool aims to reduce hallucinations and provide more accurate information when using AI for AWS-related tasks. The speakers highlight its potential to improve the reliability of AI-generated code and documentation for AWS services.

36:14 - Social engineering attack on AWS domain registration

A software company called Kodex experienced an outage due to a social engineering attack that targeted their domain registration through AWS. The speakers discuss that while the attack occurred through AWS's systems, it's not entirely fair to blame AWS as it was a human vulnerability rather than a technical exploit. They suggest that AWS and other providers may need to tighten verification policies for domain management, especially for high-profile targets.

In Season 4, Episode 37, Karl and Jon welcome AWS Community Builder Mahmoud Khatib to the show. Together, they dive into some of the latest AWS announcements, including Amazon EC2 Auto Scaling’s new support for forced cancellation of instance refreshes, Amazon RDS enabling cross-region and cross-account snapshot copy, and the expansion of AWS Organizations service control policies to cover the full IAM language. They also explore how log management can be simplified through Amazon CloudWatch centralization, and reflect on AWS being named a leader in the 2023 Gartner Magic Quadrant for AI code assistance. Somewhere along the way, the conversation takes an entertaining detour into the world of tech relics, from floppy disks to Amazon Dash Buttons.

09:19 - EC2 Auto Scaling Forced Cancellation

This new feature allows for immediate cancellation of instance refreshes, which is particularly useful when pushing out bad updates. Previously, users had to wait for in-progress refreshes to finish before canceling, potentially causing more issues. The new feature enables instant cancellation, reducing stress for on-call engineers.

15:52 - RDS Cross-Region and Cross-Account Snapshot Copy

This feature simplifies the process of copying RDS snapshots between regions and accounts. Previously, it required two steps, but now it can be done in one command. This saves time and potentially reduces costs associated with orphaned snapshots.

21:13 - AWS Organizations Service Control Policy

The update allows for full IAM language support in service control policies (SCPs). This enables more granular control and simplifies policy management, potentially replacing some use cases for permissions boundaries. The speakers were particularly excited about the improved wildcard support and the ability to centralize policies.

30:27 - CloudWatch Logs Centralization

This new feature simplifies log management by allowing easier centralization of logs from multiple accounts. It's particularly useful for organizations with multiple single-tenant applications or those needing to collect logs from customer accounts. The process is now simpler, requiring fewer steps and potentially reducing costs.

37:51 - AWS in Gartner Magic Quadrant for AI Code Assistance

AWS was named a leader in the 2023 Gartner Magic Quadrant for AI code assistance. The speakers discussed their experiences with AWS's AI coding tools, including Q Developer and Kiro, noting significant improvements in recent versions. They also expressed some skepticism about the exclusion of certain tools from the Magic Quadrant.

You can connect with Mahmoud online: LinkedIn: https://www.linkedin.com/in/mahmoud-khatib-45900052/ Medium: https://medium.com/@khatib.edge ResearchGate: https://www.researchgate.net/profile/Mahmoud-Khatib-2

In Season 4, Episode 36, Karl and Jon welcome AWS Community Builder Neeraj Sharma to the show. Their conversation covers a wide range of updates and insights, including AWS Organizations’ new Account State Information feature, the migration from Claude 3.5 Sonnet to Claude 4 Sonnet on Amazon Bedrock, and the introduction of Amazon GuardDuty’s new protection plans with extended threat detection capabilities. They also explore the transition from AWS CodeDeploy to Amazon ECS for blue-green deployments, while reflecting on AWS being named a leader in Gartner’s Magic Quadrant for cloud-native application platforms and container management. And, in true podcast style, we also discover a surprising fact: Jon’s forearm happens to be the perfect diameter for fitting a standard extractor fan vent.

09:29 - AWS Organizations' Account State Information

AWS Organizations has introduced new, more granular account states to improve account lifecycle management. This update provides clearer information about account status, including pending activation, active, suspended, pending closure, closed, and post-closure periods. The new states offer better visibility into account status, potentially improving compliance and automation workflows.

16:52 -Migrating from Claude 3.5 Sonnet to Claude 4 Sonnet

Amazon Bedrock is deprecating Claude 3.5 Sonnet and encouraging users to migrate to Claude 4 Sonnet. The migration process involves more than just changing the model flag, requiring code changes and possibly prompt engineering adjustments. The article discusses the challenges of this migration, including potential increases in token usage and changes in output format.

23:52 - Amazon GuardDuty Protection Plans

AWS has introduced new protection plans for GuardDuty, including options for S3, EKS, EC2, RDS, and Lambda. These plans offer extended threat detection capabilities and are designed to improve security monitoring across various AWS services. The article discusses the benefits and potential concerns of these new features, including automatic enablement and cost considerations.

30:47 - Migrating from AWS CodeDeploy to Amazon ECS for Blue-Green Deployments

AWS is recommending users migrate from CodeDeploy to native ECS blue-green deployments for container workloads. This change simplifies the deployment process by eliminating the need for separate appspec files and reducing complexity. The article discusses the benefits of this migration and the continuing relevance of CodeDeploy for other deployment scenarios.

37:33 - AWS in Gartner's Magic Quadrant

AWS has been named a leader in Gartner's Magic Quadrant for cloud-native application platforms and container management. The discussion covers AWS's position relative to other cloud providers like Microsoft Azure and Google Cloud, and the implications of these rankings for enterprise decision-making and cloud strategy.

In Season 4, Episode 35, Karl and Jon dive into AWS Budget Controls with automated actions, explore DMS Data Resync for seamless database migrations, and look at CloudFront’s new support for IPv6 origins. They also compare Fargate and ECS for container orchestration and discuss the shift from the SysOps Admin exam to the new Cloud Ops Engineer certification. And after hearing Jon's DIY plans, thankfully Karl failes to recall his ex tractor fan joke, as it is rather long!

02:13 - AWS Budget Controls

AWS introduced a new solution for automating actions based on budget alerts. This architecture uses various AWS services like Config, EventBridge, DynamoDB, and Step Functions to automatically manage cloud costs. Users can set up actions to inform, stop, or terminate resources based on budget thresholds for specific services like EC2, SageMaker, Aurora, and OpenSearch.

10:01 - AWS DMS Data Resync

AWS Database Migration Service (DMS) now offers a Data Resync feature for certain database engines, excluding MySQL. This feature addresses the challenge of maintaining data consistency during migrations, especially for large databases. It allows for quicker resynchronization of data without the need for a full dump and restore, which can be time-consuming for large datasets.

15:28 - Amazon CloudFront IPv6 Support

Amazon CloudFront now supports IPv6 origins, enabling end-to-end IPv6 delivery. This update offers benefits such as non-NAT operation, lower latency, and higher connection scalability. It's particularly relevant for mobile-first markets where IPv6 adoption is high. The change may also lead to cost savings as IPv6 traffic is generally cheaper than IPv4.

19:35 - Fargate vs ECS

The article discusses the differences between AWS Fargate and Amazon ECS (Elastic Container Service). It clarifies that Fargate is a serverless compute engine for containers, while ECS is a container orchestration service. The comparison aims to address confusion among users about how these services interact and their respective roles in container deployment.

22:51 - AWS SysOps to Cloud Ops Exam Evolution

AWS has renamed and updated the SysOps Administrator Associate exam to the Cloud Ops Engineer Associate exam. The new exam (SOA-C03) maintains similar content to its predecessor but includes some reorganization of topics and the addition of newer AWS services. The article discusses the implications for certification holders and the evolving nature of cloud operations roles.

In Season 4, Episode 34, Karl and Jon sit down with AWS Community Builder and Software Engineer Iyanuoluwa Ajao for a fast-paced chat on the latest in cloud and beyond. They cover the new AWS Budgets features for cross-account cost visibility, smart ways to optimize Amazon RDS and Aurora costs with Compute Optimizer, and advanced Graviton adoption strategies across regions. The conversation also dives into how AWS scaled to meet the demands of Prime Day 2025, and AWS CEO Matt Garman’s comments on why AI won’t replace junior developers. As always, the hosts veer into lighter territory—this time reminiscing about vintage computers, classic consoles, and coding in Basic.

Check out Iyanuoluwa’s blog The Rise of AI, and don’t miss Retro Reset, a tech charity close to Jon’s heart.

05:09 - AWS Budgets improvements

AWS has introduced cross-account cost visibility in AWS Budgets, allowing users to view budgets across multiple accounts within an organization. This feature is particularly useful for managed service providers and large organizations with multiple business units. It enables teams to have a consolidated view of budgets relevant to their specific areas without needing access to the entire organization.

09:23 - Optimizing RDS and Aurora with AWS Compute Optimizer

AWS Compute Optimizer now supports optimization recommendations for Amazon RDS and Aurora databases. This tool helps identify over-provisioned resources, unused instances, and opportunities for right-sizing. The article discusses the importance of database optimization due to its significant impact on overall cloud spending and provides guidance on using Compute Optimizer for databases.

16:04 - Advanced AWS Graviton adoption strategies

The article discusses strategies for implementing AWS Graviton across different AWS regions. It highlights the importance of considering regional differences in instance availability and suggests using mixed instances and instance requirements rather than specific instance types. The article emphasizes the benefits of Graviton for price and performance optimization but notes that some workloads may still require x86 chips.

22:07 - AWS services scaling for Prime Day 2025

AWS shared statistics on how their services scaled to handle Amazon's Prime Day 2025. Notable figures include deploying over 87,000 Inferentia and Trainium chips for Amazon Rufus, powering more than 40% of Amazon.com using Graviton, and processing 1.5 quadrillion daily requests on Amazon ElastiCache. This article demonstrates AWS's ability to handle massive scale and serves as a marketing tool for potential customers.

29:56 - AWS CEO on AI and junior developers

Matt Garman, CEO of AWS, stated that junior developers are not at risk of being replaced by AI. The discussion touched on the importance of understanding code versus relying solely on AI-generated solutions. The speakers emphasized the continued need for human developers, especially for smaller companies, and cautioned against over-reliance on AI in software development.