Multi-DC Cloud Foundry: What, Why and How?
summary
Cloud Foundry is stateful so CAP theorem applies: three concrete multi‑DC choices emerge — single‑site (simplest, loses CF if site dies), split control plane/runtime (apps keep running but cannot change state during a partition), or multiple independent CFs (high availability but hard consistency and rollout coordination). Practical tools demonstrated: CF Plex to run CF CLI commands across controllers, Concourse for CI-driven app pushes, the Converger for declarative org/space/service convergence, and deploy tools (including Deployed and Spinnaker) for cross‑DC app rollouts and rollbacks.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:44Story Time
- 5:33Cap Theorem
- 6:58Stateful Cloud Foundry
- 15:02CF Plex
- 18:02Concourse
- 22:39Deployment
- 25:43Questions
related talks
21mA Roadmap to Active Multi Cloud
31mPut Your Money Where Your CF Is
15mThe Cloud Foundry Renaissance
01h 03mDeep Dive into Pivotal Cloud Foundry 2.0
33mCloud Foundry Weekly: Live at CF Day Germany
36mMulti-Cloud with Cloud Foundry: Making It Real - Dmitry Bakaleinik, SAP & Sean McKenna, Microsoft
40mThe Rocket Science of Deploying Cloud Foundry in Multi-Cloud Environments
01mIntroduction: Cloud Foundry