Replatform to Amazon EKS
summary
A public-sector team replatforms a legacy three-tier estate into Amazon EKS, replacing on‑prem OKD and cutting 12+ clusters and 30+ accounts into a shared multi‑tenant platform. Infrastructure as code with Terraform and Ansible, Helm, EFS and managed node groups enable faster deploys (from a day to ~2 hours), 100% uptime for four months, a 20% ops cost reduction and reclaiming at least one FTE, while using IRSA to grant secure cross‑account pod access to S3, DynamoDB and KMS.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 0:19Agenda
- 0:48Kubernetes
- 2:23What is EKS
- 3:42EKS features
- 5:15Where does EKS cluster run
- 5:44Landing Zone
- 8:00IAM Roles
- 9:15IAM Roles for Service Accounts
- 10:04Jonathan Oliver
- 10:57Landshare Systems
- 12:18Technology Branch
- 16:30Replatform to the Cloud
- 20:46Deployment Architecture
- 24:01Sprawl
- 25:07Consolidated Infrastructure
- 28:24Multitenant Needs
- 30:54Architecture
- 38:19Summary
related talks
49mHardware to AWS, EC2 to EKS: Two Years, Two Epic Migrations
36mMigrating from kOps to EKS with Minimal Developer Involvement
23mParadox's EKS Migration
17mKeynote: Kubernetes: Lessons from the Trenches
28mBuilding K8s on AWS & Enabling Developers to Deliver. What to Do Better - Greg Sansom & Fahad Tariq
29mFrom EC2 to Kubernetes Takeaways from a Brownfield Migration
11mAmazon EKS Roadmap
15mBest practices for platform teams to streamline Kubernetes operations