Rook: Intro and Deep Dive With Ceph
summary
Rook automates Ceph on Kubernetes and is shown running a 10-region deployment with ~251 storage nodes and 2,800 OSDs providing 37 PB raw capacity today and a target of 120 PB; performance tests hit safe saturation at ~170k GETs so the operational sweet spot is 90k safe throughput. Demonstrated features: external-cluster consumption, RBD mirroring, NVMe‑over‑fabric access, RBD QoS, and seamless upgrades across Ceph releases (Squid/Tentacle/Umbrella).
AI-written summary · about
chapters (9)
- 0:00Welcome
- 1:24Today's agenda
- 2:30What is Ceph
- 5:40Rook components
- 10:40Ceph status and future
- 15:38Advanced Rook features
- 19:54SAP case study
- 27:41Conclusion
- 28:44Q&A
related talks
31mRook: Intro and Deep Dive With Ceph Storage
27mRook: Intro and Ceph Deep Dive
26mRook: Intro and Deep Dive with Ce...
30mRook: Intro and Ceph Deep Dive
30mRook: Intro and Ceph Deep Dive
26mRook: Intro And Deep Dive With Ceph Storage
12mRook: Why Would You Ever Deploy Ceph Inside Kubernetes?
19mUsing Rook in Production on National Research Platform
