A reliable distributed storage system with the object, file system, and block device interfaces.
Our service is based on Ceph, which is a decentralized cluster consisting of multiple server nodes. Ceph ensures data replication within the cluster, as well as even distribution of the load on the nodes. In case of disk or node failure, Ceph provides uninterruptable operation and data availability.
You can work with storage at the object level by means of RESTful API. By sending specific REST-requests over HTTP(S), users or applications can store, read, delete objects, and set access rights for the objects.
• S3-compatible provides an interface that is compatible with a large subset of the Amazon S3 RESTful API;
• Swift-compatible provides an interface that is compatible with a large subset of the OpenStack Swift API.
With block storage interface, a part of Ceph storage (image) can be mounted to the client machine as a network block device. After Ceph storage is mounted, you can create partitions and file systems on it or use it as a raw block device. Block storage is provided via iSCSI and ceph rdb protocols.
• native support of iSCSI protocol on Windows and Linux;
• native support of rdb protocol on Linux;
• resizable images;
• image copy or rename;
• integration with QEMU KVM;
• integration with Kubernates/Openshift;
Storage also offers a traditional file system (CephFS) interface with POSIX semantics. As CephFS support is integrated into Linux kernel, no additional client customization is required.
• applications that use POSIX file systems can use Ceph FS with no additional customization instant snapshot creation at directory level
• instant snapshot creation at directory level
• kernel and fuse modes of operation