Fast-track Kubernetes and platform rebuilds with observability
summary
CLEAR rebuilds its biometric matching engine onto Kubernetes and microservices on AWS EC2/EKS with near-zero downtime, completing the platform migration in six weeks. Tests run with Locust and Jenkins push client-side metrics into SignalFx, letting teams compare production traffic, latency and throughput. Universal Scalability Law modeling identifies contention and coherency limits, predicts max sustainable load, and informs tradeoffs (CPU vs. more instances). Outcome: faster development cycles, calibrated production-grade tests, measurable capacity headroom, and targeted cost/performance tuning.
AI-written summary · about
chapters (15)
- 0:00Intro
- 2:07Our Challenge
- 9:35Overconfidence
- 13:10Tools
- 15:46Observe and Baseline
- 17:38Build Some Tests
- 19:42Run Some Tests
- 26:17What Does It Show You?
- 29:34Experiments
- 36:24PDO Performance Model
- 39:39Explore Opportunities
- 42:50Faster Development
- 44:41Faster Testing
- 45:28Better Operations
- 46:46Better Planning
related talks
23mScaling Kubernetes in Regulated Environments Through Platform APIs
39mStrangling the Legacy with Kubernetes
56mBest practices for optimizing Kubernetes applications on AWS
37mRunning Kubernetes in a Manufacturing Line
23mMore Power, Less Pain: Building an Internal Platform with CNCF Tools
28m03 Kubernetes Evolution, from the Hard Way to Managed Multi-cloud
26mMulti-Tenant vs Micro-Clusters in Kubernetes-Driven Platform Engineering
36mBuilding a CI/CD driven infrastructure for managing Kubernetes clusters