Starring the eleventh annual Red Hat Summit in Boston this week are new releases of the company’s core software-defined storage technologies that promise to bring the performance and efficiency of commodity-based capacity much closer up to par with the standards of the traditional enterprise. The biggest improvements are to its block and object service.

Red Hat Ceph 1.3, which is based on technology that the open-source giant obtained through its nine-digit acquisition of Inktank Inc. last year, provides a storage backend for OpenStack clusters that now ships with a embedded web server to help speed up requests for the objects kept inside. And manipulating data stored in its native block form has also been made faster with the same stroke.

The impact of the performance optimizations introduced in the release extends beyond pulling information to protecting it as well, with the “scrubbing” function responsible for ensuring that all the data placed into the environment is accounted for having been enhanced with a scheduler in conjunction. That kills two birds with one stone.

Administrators can use the feature to prevent automated integrity checks from running during peak hours when the added overhead from the procedure may hinder use experience. That optimizes performance while allowing for more efficient data protection, a focus that is also evident in the new release of Gluster that made its debut alongside Ceph Storage 1.3

The file service, which rounds out the three storage formats that Red Hat supports, has been augmented with the addition of erasure coding technology that makes it possible to reconstruct data on a failed disk based on the remaining snippets in the rest of the cluster. That’s essential for the kind of massively distributed environments in which OpenStack is typically deployed, where failed nodes are usually disposed instead of repaired.

The update also provides administrators with more control what data is kept in which part of their implementations via a complementary tiering function that can automatically relegate information to the most appropriate storage medium based on access patterns. Frequently used records are stored on speedy flash while archives get written to something slower and more affordable.

Rounding out the new release of Gluster is a bit-rot detector that scans for the natural degradation of storage media over time to help organizations combat the resulting data corruption. Added up with the other features introduced as part of the launch, that makes Red Hat’s stack much more appealing for powering mission-critical enterprise workloads that require a combination of efficiency and reliability, but there’s still a long way to go until software-defined storage is truly ready to hit prime time.

Photo via Bob Mical