Red Hat Storage Server (RHSS) storage software has been upgraded to allow scalability to a potential 19PB, and has added features including snapshots and compatibility with the Hadoop Distributed File System (HDFS).
Version three of RHSS sees the number of drives per server increased to 60 from 36, and the number of servers that can be clustered together doubled to 128 from 64. That means customers can potentially build 1PB of storage from three server nodes or 19PB from the maximum number of nodes.
RHSS is storage software that can be installed on any x86 server with commodity drives. It is based on Red Hat Enterprise Linux (RHEL) 6 and the open-source GlusterFS 3.6 file system. Red Hat acquired Gluster at the end of 2011 and began to release products in 2012.
Such storage software forms part of what has more recently been dubbed software-defined storage. The chief benefits are that customers can build their own storage arrays with off-the-shelf components that are substantially cheaper than buying a storage system from the big six storage suppliers.
More on open-source storage
RHSS can be deployed to an existing instance of RHEL. Other features in version three include an HDFS plugin for analytics operations, up to 256 snapshots per volume with user-serviceable snapshots as a beta feature, and integration with Nagios open-source monitoring software.
Vice-president and general manager for storage and big data at Red Hat Ranga Rangachari said RHSS is aimed at workloads such as big data, security logging with software like Splunk, and enterprise collaboration.
“It’s not a tailored product, but there will be reference architectures available for different use cases via the channel,” he said.
RHSS 3 provides file-based access of storage to block and object storage capabilities alongside Red Hat's recently acquired and updated Inktank Ceph Enterprise, which "provides a complete spectrum of services", according to Rangachari.
In July, Red Hat upgraded the OpenStack storage software distribution Inktank Ceph Enterprise (ICE) to version 1.2. This saw the addition of erasure coding data protection to provide cost-efficient backup and archive use, as well as tiered storage functionality.