Ceph存储集群

CEPH STORAGE CLUSTER

 Ceph Storage Cluster是部署的基础。基于RADOS，Ceph存储集群基于两类进程：

• Ceph OSD Daemon（OSD）：将数据作为对象存储在存储节点
• Ceph Monitor：维护一个cluster map的副本

一个Ceph存储集群可能包含成千上万个存储节点。一个最小的系统则包含最少一个Ceph Monitor和进行数据复制的两个OSD。

Ceph文件系统、Ceph对象存储、Ceph块设备都痛Ceph存储集群读取或者写入数据。

配置和部署 Ceph存储集群有一些必要的设置，但是大多数设置有默认值。一个典型的部署是使用一个部署工具定义一个集群和引导一个monitor。 部署工具ceph-deploy参考 Deployment 配置 Disks and Filesystems Configuring Ceph Network Settings Auth Settings Monitor Settings Heartbeat Settings OSD Settings Filestore Settings Key/Value Store Settings Journal Settings Pool, PG & CRUSH Settings Messaging Settings General Settings Deployment Preflight Checklist Install Ceph Create a Cluster Add/Remove Monitor(s) Key Management Add/Remove OSD(s) Add/Remove MDS(s) Purge Hosts Admin Tasks

操作 一旦你部署完毕一个Ceph集群，你就可以对集群进行操作了。 Operations Operating a Cluster Monitoring a Cluster Monitoring OSDs and PGs User Management Data Placement Overview Pools Erasure code Cache Tiering Placement Groups CRUSH Maps Adding/Removing OSDs Adding/Removing Monitors Command Reference The Ceph Community Troubleshooting Monitors Troubleshooting OSDs Troubleshooting PGs Logging and Debugging CPU Profiling Memory Profiling Man Pages

APIS Most Ceph deployments use Ceph Block Devices, Ceph Object Storage and/or theCeph Filesystem. You may also develop applications that talk directly to the Ceph Storage Cluster. APIs Introduction to librados librados (C) librados (C++) librados (Python)