CERN OpenStack User Story
summary
CERN deploys OpenStack into production with Grizzly production July 2013 and runs four Icehouse clouds that present as a single pool while actually scaling via Nova cells. The on‑site cloud reaches 75,000 cores today, adds 45,000 cores in a second room underground, and expects ~150,000 cores by Q1 2015; tooling uses puppet, Cabana, elastic search, kibana and Seth storage; all community‑relevant code is upstreamed and the cloud is actively serving physicists.
AI-written summary · about
chapters (10)
- 0:00Introduction to CERN
- 1:28Physics challenges
- 4:05The Large Hadron Collider
- 6:15CERN computing center
- 8:33Expanding IT infrastructure
- 10:09Adopting open-source
- 11:43OpenStack implementation
- 14:07Cloud federation and culture
- 15:58Overcoming resistance
- 18:22Conclusion
related talks
19m10 years of OpenStack at CERN (From 0 to 300k cores)
18mOpenStack Playing for an Open World
18mScaling Bare Metal Provisioning with Nova and Ironic at CERN
30mOpenStack Works, so now what?
44mLINE's Journey; Road to 4 Million Cores in t...
38mHow OpenStack became boring (and successful)
49mIntroduction to OpenStack
54mOpenStack Development: The Infrastructure Behind the Infrastructure