Red Hat Inc. has just rolled out an important new release of its software-defined object storage software Ceph, which it says will be the “foundation for the next long-term stable release”.

The aim of Ceph is to create a totally reliable storage system that has no single point of failure. This is done by spreading data across a single distributed computer cluster, providing interfaces for object-, block- and file-level storage, and aspiring to scale to exabytes of storage. Being open-source, the project and its related software are also entirely free to download and use.

There’s a few nifty new features in the release, called “Jewel”, according to the notes in Ceph’s latest blog post. The headline act is almost certainly the first feature-complete and stable release of the Ceph filesystem, which offers goodies like snapshots, disaster recovery tools, volume repair and experimental support for multiple Ceph files systems spread across a single cluster.

There’s also greater compatibility with OpenStack’s Swift storage API, while another important change means that systemd manages the Ceph process, while the Ceph daemon runs as a user called “ceph’ instead of root. Also, SHEC erasure coding has been upgraded from its old experimental status.

Ceph has also been trying to standardize on C++11, which means support for some older operating systems has been withdrawn. That means no more support for CentOS 6, Debian Wheezy 7.x, Ubuntu Precise 12.04, and others. Instead, you’ll need to be running CentOS 7.x, Debian Jessie 8.x Ubuntu Trusty 14.04, Fedora 22 or aarch64/arm64 or later versions of those OSs.

The official blog post from Ceph detailing the release is a little odd, because it states that Jewel is a “foundation” for the next long-term release (LTS). However, in the project’s release schedule, Jewel is officially listed as having LTS status.

Photo Credit: RichardBowen via Compfight cc