Gluster, a maker of open-source storage software, announced GlusterFS 3.3 beta 2 today. The major addition this update brings is the introduction of Hadoop support, a move that likely made Gluster’s platform suddenly much more interesting for a lot of new clients.

“With this new feature, GlusterFS is the first open source, POSIX-compliant file and object storage solution that works with Hadoop. Compatible with existing MapReduce-based applications and with the ability to co-exist with HDFS, this new functionality opens up data within Hadoop deployments to any file- or object-based application.”

Version 3.3 beta 2 comes with N-way replication, a combo of synchronous and asynchronous replication, so that data remains synced regardless of latency or potential network issues.  The platform also features failover, no single point of failure, and enables clients to run jobs without the 64MB block limit and 16 TB volume size cap.

Lastly, Gluster eliminates the need for a metadata or name server thus allowing faster access to data, and offers POSIX-compliant NAS and object access to files in a Hadoop deployment.

Gluster’s dev team has been working on GlusterFS intensively this year.  The company unveiled GlusterFS 3.3 beta 1 only a month ago, which introduced file and object data storage integration.  This helps to better integrate object storage with legacy applications, as well as to scale NAS to the public cloud.

Gluster is trying to expand its ecosystem via integration with other open-source cloud and data offerings.  The addition of Hadoop support is the latest move, followed by a contribution to OpenStack initiative s week later.

The code, dubbed Gluster Connector, connects GlusterFS to OpenStack, and allows users to add more VMs and allow their movement within the deployment.  This means files no longer need to be stored locally, and they can be more readily shared in the cloud.