Using Kubernetes for Handling Second Screen Experience of European TV Show
summary
Describes delivering a highly scalable second-screen application for Danish TV show “Fuld Plade” using Go (GoLang) services deployed on a Kubernetes cluster to handle live spikes (up to 100x load during the hour-long show). Explains how they used Prometheus for performance monitoring and gaining operational insights into the application under heavy, time-bound load.
AI-written summary · about
chapters (19)
- 0:00Start
- 0:23About us
- 1:16About Peytz & Co
- 4:03Challenge: Timing!
- 6:08Application architecture
- 8:42Application technologies
- 9:19Scaling and orchestration
- 10:49Deployment, Cl and CD
- 12:57The services around Kubernetes
- 14:25KOPS: Installation
- 14:45KOPS: Create cluster
- 15:38KOPS: Edit ig nodes
- 17:08KOPS: Validate cluster
- 17:22Monitoring
- 19:59Performancetest: LOCUST
- 20:38LOCUST: Fast Forward
- 23:37Evaluation
- 24:17Record breaking
- 24:57Conclusion
related talks
21mKeynote: Pushing the Limits of Kubernetes with Game of Thrones
23mHighly Available Live Encoding Using Go and Kubernetes
35mProgramming Kubernetes with the Go SDK [I]
19mPains and gains of Kubernetes for Developers, story from experience
22mKeynote: Kubernetes: As Seen On TV
02h 18mDevelop and Deploy to Kubernetes like a Googler
40mCase Study: Kubernetes at Comcast
42mDelivering TV Everywhere with Cloud Native Solutions