Messages in this thread | | | Date | Wed, 7 May 2008 15:32:57 -0700 (PDT) | From | Sage Weil <> | Subject | [ANNOUNCE] Ceph distributed file system v0.2 |
| |
Hello everyone,
Ceph is a distributed file system designed for performance, reliability, and scalability. Basic features include:
* POSIX semantics * Seamless scaling from 1 to many thousands of nodes * No single point of failure * N-way replication of data across storage nodes * Fast recovery from node failures * Automatic rebalancing of data on node addition/removal * Easy deployment: most FS components are userspace daemons * Linux kernel client, FUSE-based client, and user library
Notable in this release is a reasonably stable Linux kernel client. It can extract and compile a kernel, and passes all tests in the fstest POSIX regression test suite posted a few weeks back. It has stabilized to the point where it could use some broader testing and code review.
More info at
http://ceph.newdream.net
Source code at
git://ceph.newdream.net/ceph.git http://ceph.newdream.net/git
Since v0.1:
* fully functional (and reasonably stable) kernel client * NFS re-export of a ceph client mount * client metadata leases to keep client cache coherent * crushtool for managing storage cluster topology * improved support for storage cluster expansion * some new tools for mkfs and management * lots and lots of bug fixes...
Planned for v0.3:
* xattrs * hardening distributed failure recovery * large directory support (in client) * recursive mtime and file size accounting
Some key things on the (medium- to long-term) roadmap:
* locks * quotas * btrfs for local object storage on storage nodes * directory-granularity snapshots * content-addressible storage * strong security
sage
| |