How migrating from ECS to EKS supercharged our infrastructure
summary
Migrated 200+ microservices from ECS to EKS, replacing manual ECS service definitions with declarative Helm charts and Argo CD; node provisioning moves to Cluster Autoscaler/Karpenter enabling aggressive scale-down and almost 100% of GRS on Spot instances. Cluster-level utilization rises from ~36% CPU and 8% memory to ~60% each, cutting costs and reducing manual intervention. Visibility and security use Falco and Wazuh for network/host auditing, metrics-server and APM integrations for app-level signals, and canary/log collection workflows are automated into CI/CD. Helm charts, Argo CD, Karpenter, Falco + Wazuh
AI-written summary · about
related talks
36mMigrating from kOps to EKS with Minimal Developer Involvement
49mHardware to AWS, EC2 to EKS: Two Years, Two Epic Migrations
47mPostmortem: A Modern Approach to Managing Kubernetes Clusters
47mBuilding sustainable architectures with Kubernetes
23mParadox's EKS Migration
01h 02mLearnings from the Enterprise-level Kubernetes transformation at Elkjøp Nordic
25mHow to take your Kubernetes cluster to next level with Karpenter, Ray and Amazon EKS?
43mHow a Unicorn Migrated From an In-House Solution to Helm