Training Agenda

Ceph Storage

Ceph is a distributed storage system that provides object storage (RADOS), block storage (RBD), and filesystem storage (CephFS) from a single unified cluster. It scales from terabytes to exabytes, handles hardware failures automatically through its CRUSH algorithm, and is the storage foundation behind OpenStack and many Kubernetes deployments — often managed via the Rook operator.

2 days On-site, remote, or hybrid Up to 20 participants German or English
What We Cover
Unified distributed storage — block, filesystem, and object for Kubernetes and beyond
Day 1

Architecture, CRUSH & Cluster Operations

  • RADOS architecture: Monitors (quorum and map distribution), OSDs (data and metadata), Managers (dashboard, metrics), MDSs (CephFS)
  • CRUSH algorithm: placement group mapping, CRUSH maps (buckets, rules, host vs rack failure domains), custom CRUSH rules
  • Replication vs erasure coding: 3-way replication for performance, erasure coding (k+m) for capacity efficiency — trade-offs
  • CephX authentication: key management, client capability strings (mon/osd), ceph auth commands
  • Cluster deployment with cephadm: bootstrap, adding OSDs, service specs in YAML, ceph orch commands
  • OSD management: adding and removing OSDs, rebalancing (osd_recovery_*), replacing failed disks safely
  • Cluster health: ceph status, ceph health detail, understanding HEALTH_WARN vs HEALTH_ERR, alert rules
Day 2

Storage Interfaces, Kubernetes & Rook-Ceph

  • RBD block storage: pool creation, image management (create/resize/snapshot/clone), rbd map and mount
  • RBD as Kubernetes PVs: ceph-csi rbd driver, StorageClass configuration, dynamic provisioning, snapshot support
  • CephFS shared filesystem: MDS active/standby configuration, volume and subvolume management, quotas
  • CephFS as Kubernetes RWX volumes: ceph-csi cephfs driver, ReadWriteMany PVCs for multi-pod shared access
  • RADOS Gateway (RGW): S3-compatible API, bucket and user management, bucket policies, quota enforcement
  • Multi-site RGW replication: zone groups and zones for geo-redundant object storage
  • Rook-Ceph operator: CephCluster CRD, CephBlockPool, CephFilesystem, CephObjectStore, StorageClass provisioning
Learning Outcomes
What your team walks away with

Participants can deploy and operate a Ceph cluster, provision all three storage types for Kubernetes workloads via CSI drivers, and handle routine maintenance including disk replacement and capacity expansion.

Book the Ceph Storage training

Reach out to schedule a session for your team — remote, on-site, or hybrid, in German or English.

Get in touch