Current location - Education and Training Encyclopedia - Graduation thesis - Ceph stores the difference between the layer of ceph cluster and RBD cache.
Ceph stores the difference between the layer of ceph cluster and RBD cache.
Ceph is a high-performance and extensible single-point distributed file storage system. Paper development based on Sage A. Weil mainly provides the following three storage services:

Object storage, compatible with Amazon S3 and OpenStack Swift, can access or store data in the form of objects by using Ceph library, C, C++, Java, Python and PHP codes, or through Restful gateway.

As a block device, block storage is directly mounted like a hard disk.

File systems mounted like network file systems are compatible with POSIX interfaces.

The structure of Ceph, object storage is provided by LIBRADOS and RADOSGW, block storage is provided by RBD, and file system is provided by CEPH FS. RADOSGW, RBD and CEPHFS all need to call the interface of LIBRADOS, and finally they are all stored in RADOS in the form of objects.

The nodes of Ceph cluster have three roles:

Monitor, which monitors the health of the cluster and sends the latest CRUSH map (including the current network topology) to the client.

OSD, maintaining objects on nodes, responding to client requests and synchronizing with other OSD nodes.

MDS provides the metadata of the file. If you don't use CephFS, you don't need to install it!