Darryl R.’s Post

If you’ve chosen to run you application in Kubernetes on AWS using the Elastic Kubernetes Service (EKS) you will want to setup monitoring tools to keep an eye on everything. There are lots of open source tools available and most can be setup rather easily in your cluster. You will have to maintain these and mange the compute and storage for them. If you want to use these tools (like Grafana and Prometheus) you can also take advantage of the managed versions of them and leave most of the work to AWS. You will have a cost but it will be up to you to decide if it’s worth it. This article from Siva Guruvareddiar and Michael Hausenblas shows how to set these up with your EKS cluster. https://lnkd.in/eWNSnNxF

Always great to see new ways to make architectures both more resilient and more performant.

New Amazon Web Services (AWS) Compute Blog post today on implementing multi-Region failover with Amazon API Gateway! This implementation is unique in enabling independent failover for discrete microservices deployed behind a shared public API. This can be a powerful way for enterprises to raise the bar on resiliency at scale. Thanks to Marcos Ortiz, Khubyar Behramsha, and Chris Munns for the great work here. https://lnkd.in/gaKjhAHU

AWS recently introduced a new feature that makes EKS control plane metrics available. These metrics, previously inaccessible, will now be rolled out to all clusters, including existing ones on version 1.28 or older. It'll be interesting to see metrics like those from the scheduler and API. While EKS is still a managed service, and we shouldn't expect too much detail, some of these hidden metrics could help with troubleshooting. https://lnkd.in/ekbtKVPB

Just published a new blog post along with Imaya Kumar Jagannathan discussing how to leverage KEDA and Amazon Managed Service for Prometheus to automatically scale your Kubernetes workloads! 🔥💻 KEDA enables event-driven autoscaling based on various event sources, including Prometheus metrics from AWS. This approach allows for granular scaling decisions to optimize resource utilization in your Kubernetes environment. Check out the full post to learn about the technical details and implementation steps for this powerful autoscaling approach on AWS. 🚀 Read the post here: https://lnkd.in/gJjHyFDK #blog

Mohammed A. and I published an Amazon MQ for ActiveMQ benchmarking blog post on the AWS Compute blog to allow customers easily evaluate and perform benchmarks during their migration or adoption of Amazon MQ for ActiveMQ. In this blog post, you will learn how to 1/ build a docker image baked in Maven 2 plugin (using the open wire protocol) 2/ Deploy benchmark infrastructure using AWS CDK for python. 3/ Benchmark against Amazon MQ instances using ECS exec and your docker image. 🔗 Read more here: https://lnkd.in/e46EfRJX

Keeping track of resource usage in your Kubernetes clusters is key to ensure you're not wasting a lot of $$$s. There are a number of approaches to deal with this but one solution that works well with AWS and Elastic Kubernetes Service (EKS) clusters is KubeCost. KubeCost scans the resources in your clusters and offers a nice front-end UI showing the status. It offers suggestions on optimizations you can make. It also includes dollar amounts with estimates of savings that you could have by making changes. Below, Mahendran Selvakumar shows how to setup KubeCost in your EKS clusters and start using it. https://lnkd.in/eR-nNDyP

Advanced AWS ALB routing to PODs running across EKS clusters. Background 1) Each AWS ALB/ELB is created with respective EKS ingress/Load Balancer type . This increase number of Load balancers if you publish multiple micro services. 2) Routing across EKS cluster needs complex routing and Load balancers on top of ALB/ELB created as part of Ingress/Load Balancer type. Solution. 1) Leverage  Kubernetes custom resource (CR) by AWS Load Balancer Controller, called TargetGroupBinding. 2) TargetGroupBinding CR binds a Kubernetes Service to a load balancer target group. This can reduce number of Load balancer for your EKS. #awscloud , #awscommunity , #awscommunitybuilders , #loadbalancing , #costoptimization https://lnkd.in/gi-MvpCT

New Blog: Optimizing Amazon MQ Throughput with ActiveMQ Mohammed A. and Enigbokan olajide share best practices for benchmarking Amazon MQ using the ActiveMQ Classic Maven Performance test plugin. Learn how to optimize performance, choose the right instance types, and configure your setup for better throughput. Get practical tips and a step-by-step guide to improve your Amazon MQ configurations. Check it out!

Get deeper visibility into your AWS serverless apps with Datadog's enhanced distributed tracing for serverless. With Datadog APM, you can instrument your serverless workloads to get deep insight into requests as they flow across your functions, containers, and other infrastructure components. Providing you with more complete end-to-end visibility into Amazon S3 and DynamoDB state changes and AWS Step Functions. Datadog #serverless #apm