Your Data Center will Implode in 30 days, a Success Story
summary
A forced 30‑day migration moves a Cloud Foundry platform and 300 applications (about 2,000 instances) plus six brokered data‑service teams, copying databases and large blobstores and standing up staging, preprod and production instances. Data services are proxied from a third site to avoid wide outages, requiring one knowledgeable member per data‑service team and two focused platform engineers; there is a 24‑hour deployment freeze and the cutover yields minutes of downtime, five troubled apps and no reported data loss with help from Stark & Wayne’s Shield.
AI-written summary · about
chapters (8)
- 0:00Introduction to the crisis
- 3:14Hiring a new director
- 5:04Planning the migration
- 10:49First migration hurdles
- 12:25Migration completion
- 15:56Post-migration cleanup
- 16:39Real story behind the talk
- 21:53Success and lessons learned
related talks
24mMigrating Large Scale CF Environments
33mCloud Foundry Weekly: Live at CF Day Germany
29mCloud Foundry Growing Pains: Lessons Learned Keeping the Largest CF Environment Alive and Kicking
25mMass migrations to the cloud
21mKeynote: Cloud Foundry at Rabobank after 10 months of real use
31mFrom a data center to a private cloud (DATEV eG)
48mCloud Foundry - your PaaS on OpenStack
50mGoing from Nothing to Production Within an Hour