If You Build It, They Will Come - a Platform Modernization Journey
Talkat Platform Engineering Day NA 2024 · on Tuesday, 2024-11-12· at 15:55· Salt Palace | Level 1 | Grand Ballroom G·
summary
Built a company-first cloud platform that replaces on-prem Docker Swarm/Rancher with Amazon EKS and a suite of tools to fix scalability, observability, security and usability. Node autoscaling and consolidation use Karpenter with taints/tolerations and multi‑AZ provisioning; cluster config backups use Velero; service mesh observability and policy enforcement use Tetrate (Istio). Result: ~90% of deployments automated, on‑call midnight kubectl restarts disappear, CI lifecycle tied to ServiceNow, cost visibility via Cloudability, and tenant isolation with namespace+policy controls.
AI-written summary · about
chapters (13)
- 0:00Introduction and context
- 1:21Identifying legacy challenges
- 1:43Infrastructure pain points
- 5:32Planning for success
- 5:47Building the platform
- 6:48Customer collaboration and tools
- 13:04Observability improvements
- 15:00Security and compliance
- 16:31Tenancy and access control
- 18:44Refining cluster permissions
- 20:42Infrastructure and scaling
- 21:36Karpenter and autoscaling
- 26:03Conclusion
related talks
34mTaking Back Control: Building Cloud-Agnostic Container Platforms for Regulated Environments
33mBlueprints of Innovation: Engineering Paved Paths for a User-friendly Developer Platform
40mScaling Startup for Success: Building modern Architectures and Developer Acceleration
09mHow migrating from ECS to EKS supercharged our infrastructure
35mNavigating the Kubernetes Odyssey: Lessons from Early Adoption and Sustained...
34mReduce Costs and Optimize on Your Cloud with Karpenter
37mBuilding a Managed Platform While Maintaining a Good Developer Experience
26mMulti-Tenant vs Micro-Clusters in Kubernetes-Driven Platform Engineering