CephFS provides a fault-tolerant, elastic, scalable open source file store that leverages the same distributed Ceph storage cluster that many already use to provide OpenStack object and block storage solutions.
We'll cover how Manila has been extended to support export of shares backed by CephFS via the widely-used NFS protocol to tenant VMs, reasons for using NFS instead of or in addition to using native CephFS as the share protocol, and current and planned approaches to solve multi-tenancy, performance, and scale-out issues.
This talk will be of immediate practical interest to cloud operators who have or are planning to use Ceph clusters for object and block storage and want to use the same cluster for shared file services as well as to those more generally interested in Manila, distributed file systems, or Ceph evolution.
- Background in manila, cephfs, and nfs ganesha
- Deployment at scale with the *native* cephfs storage back end
- Deployment with the CephNFS storage back end
- Futures including:
- active-active CephNFS gateways
- dynamic launch and scale out of CephNFS gateways per tenant
- NFS performance improvements
- file service to containers and clients outside OpenStack proper