Cloud Foundry Availability
summary
Defines platform availability by building a dependency graph of atomic services, composing serial and parallel reliabilities (example: 99.5% per component yields lower serial availability, near‑100% for parallel), and aggregating environments with weighted averages to hit targets like 99.9%. Measures come from functional checks plus Prometheus monitoring (Monit for processes) and BOSH metadata; demonstrates 3‑node replica math (need 2 of 3) and shows rolling BOSH upgrades can be arranged to avoid platformwide outages.
AI-written summary · about
chapters (8)
- 0:00Start
- 8:29Reliability Plock Diagrams
- 12:52Serial Dependency
- 20:34Weighted Averages
- 22:55Subsystems within a Cloud Foundry Environment
- 25:26How Do You Obtain Availability from Atomic Services
- 31:12Perform Availability Monitoring during Upgrades
- 35:16Platform Dependency Craft
related talks
05mFour Levels of High Availability in Cloud Foundry
43mHow Platforms Work
27mCloud Foundry in Production
31mPut Your Money Where Your CF Is
33mCloud Foundry Runtime PMC Overview - Dieu Cao, Pivotal
17mDeployments We Can Believe In
25mCloud Foundry in Simple Terms: A Beginner's Journey
21mImproving the Reliability of Cloud Foundry Applications Through Advanced Health Checks