Ceph Storage: a Caffeinated Primer
summary
Ceph is a unified object/block/file system that the talk demonstrates end-to-end: RADOS + CRUSH + PGs place data across OSDs and monitors, supports replication and erasure coding, and scales to about 100 PB for object and ~10 PB for block/file cases. The live demo builds a 3-node cluster, deploys RGW (S3), creates a bucket, provisions an RBD image and filesystem, and exposes CephFS via an NFS gateway so reads/writes show immediate coherence.
AI-written summary · about
chapters (64)
- 0:00Ceph Storage intro
- 1:04What is Ceph used for
- 3:21Software defined storage
- 4:45Ceph unified storage
- 6:11Ceph scalability & performance
- 8:18Ceph as Linux of storage
- 9:43Ceph community & open source
- 10:44Hardware needs & tradeoffs
- 12:50Ceph on non-x86
- 14:37Ceph stats & deployments
- 17:28Distributed systems in Ceph
- 18:33Ceph resiliency & durability
- 20:40Ceph performance vs safety
- 22:05Ceph upgrade method
- 23:49Ceph disaster recovery
- 25:55Ceph unified interfaces
- 26:59RADOS the reliable autonomic distributed object store
- 29:05Ceph APIs & clients
- 30:09RADOS Gateway
- 30:51RBD block storage
- 31:38CephFS file system
- 32:23RADOS components
- 34:32Ceph monitors & state
- 35:57Ceph manager
- 37:00Ceph direct data access
- 38:47Ceph peer-to-peer storage
- 40:12RADOS objects & metadata
- 41:39Placement groups
- 45:49CRUSH algorithm
- 49:19Ceph data safety
- 57:15Ceph replication & erasure
- 1:00:53Ceph pools & namespaces
- 1:02:18Ceph cluster scaling
- 1:03:22RADOS Gateway arch
- 1:06:30Ceph geo replication limits
- 1:07:36RADOS Gateway zones
- 1:10:07RADOS Gateway S3
- 1:18:10RBD fundamentals
- 1:20:14RBD snapshots & cloning
- 1:23:35RBD replication modes
- 1:26:45RBD features & integration
- 1:27:28CephFS architecture
- 1:31:39CephFS caching & performance
- 1:33:11CephFS client interfaces
- 1:35:23CephFS metadata scaling
- 1:38:11CephFS snapshots & quotas
- 1:43:52CephFS multiple filesystems
- 1:44:57CephFS data layout
- 1:46:01CephFS installation
- 1:47:07Ceph dashboard & tools
- 1:49:15Ceph install & orchestration
- 1:51:23Ceph community & releases
- 2:00:13Ceph in HPC
- 2:13:17Live Ceph demo setup
- 2:21:46Bootstrapping Ceph
- 2:26:14Deploying OSDs & monitors
- 2:31:19Ceph cluster status
- 2:34:31Deploying RADOS Gateway
- 2:36:49S3 buckets & users
- 2:40:40Ceph pool management
- 2:43:18Creating CephFS
- 2:45:08Deploying NFS Gateway
- 2:51:31Mounting CephFS
- 2:55:13Creating RBD devices
related talks
49mIntro to Ceph, the Distributed Storage System
29mCeph Operations at Scale
34mModernizing Ceph Deployments at CERN: CephFS and Object Storage Across Data Centres
38mTesting Ceph: Pains & Pleasures
26mImplementing a Resilient Ceph RGW Object Storage with Erasure Coded Data Pool as the Foundation...
29mGlobal Deduplication for Ceph
35mMaking Ceph Fast in the Face of Failure
35mKeeping Ceph RGW Object Storage Consistent