Start Here

Read the Learn pages, then deploy a test cluster under Set Up. The Look Up pages hold the numbers you need while you plan.

Learn

Read these before you install anything.

Beginner’s Guide: what Ceph is, in plain language.
The Parts of a Ceph Cluster: the daemons in a cluster, what each one does, and how many you need.
Architecture: how Ceph stores data, places it with CRUSH, and recovers from failures.

Set Up

Deploying a Single-Host Test Cluster: a working cluster on one machine, for learning and testing.
Creating and Mounting a Block Device: a first block device on the new cluster.

Look Up

Hardware Recommendations: how to choose CPUs, memory, storage devices, and networks for a cluster.
Minimum Hardware per Daemon: the smallest configuration for each daemon.
CPU and Memory Sizing: cores and RAM for each daemon, and the settings that control memory use.
Storage Devices: drive layout, HDD and SSD selection, controllers, and write caches.
Network Sizing: link speeds, replication times, bonding, and management networks.
OS Recommendations: the platforms that each Ceph release is built and tested on.
Ceph Releases: the release cycle and how long each release is maintained.
Active Releases: the releases that are maintained now, with their release notes.
Glossary: definitions of the terms used throughout this documentation.

Next Steps

Deploy a production cluster. See Using Cephadm to Deploy a New Ceph Cluster, or Installing Ceph for the other installation methods, including Rook for Kubernetes.
Set up the kind of storage that your applications need: Ceph Block Device, Ceph File System, Ceph Object Gateway, or Ceph CSI for Kubernetes.

Additional Resources

Ceph Community Channels: mailing lists, chat channels, and events.
Troubleshooting: what to check when a cluster is unhealthy.
Documenting Ceph: how to fix or improve this documentation.

Brought to you by the Ceph Foundation

The Ceph Documentation is a community resource funded and hosted by the non-profit Ceph Foundation. If you would like to support this and our other efforts, please consider joining now.