Gluster rolls in Hadoop

Gluster, developer of an open-source, distributed file system capable of scaling to several petabytes, recently released a new version of its GlusterFS object storage solution that is compatible with Apache Hadoop. GlusterFS 3.3 beta 2 is available now and will be generally available in Q4. We talked with Anand Babu (AB) Periasamy, CTO and co-founder of Gluster, about the new solution.

What’s the significance of being compatible with Apache Hadoop?

With the addition of Apache Hadoop compatibility, Gluster has opened up the data stored and analyzed in Hadoop to any file-based application without requiring a rewrite. It also allows Hadoop deployments to take advantage of the enterprise-grade file system features in GlusterFS like replication, fail-over, self-healing and more. We used the standard file system APIs available in Hadoop for seamless integration into the data processing infrastructure, providing a new storage option for Hadoop deployments. Compatible with existing MapReduce-based applications and with the ability to co-exist with HDFS, this new functionality provides new data access options for Hadoop deployments, allowing organizations to benefit from both Hadoop and GlusterFS in one deployment. 

Why is storage such an important aspect in the cloud?

Enterprises are beginning to recognize the need for better cloud storage in order to achieve optimal savings and availability from their cloud deployments. The management and monitoring of a cloud infrastructure is essential to high performance. Cloud storage plays a critical role and is a major element in simplifying enterprises’ path to the cloud while maximizing budgets.

What areas will you focus on for the next release of GlusterFS?

We have a lot of plans for GlusterFS. Requests are coming in from our customers and from our active community. We are continuing our work on making Gluster deployments even easier to manage. In future releases, we will also further our security support to facilitate integration in the enterprise infrastructure, and announce new integration with additional cloud infrastructure providers—and more.

What’s next for Gluster overall?

Gluster has a lot going on at the moment, with great market opportunity that we are eager to take advantage of. We sit at the intersection of three growing market trends: storage, open source and cloud. We want to build on our current success with more enhancements to our software-only approach to storage. With the growing confusion in the cloud industry and the constant introduction of new solutions, we will also make it easier for organizations to consolidate their storage into one platform running on any infrastructure on-premise, virtual, private or public cloud. Additionally, we will continue to work closely with our open source community. Finally, we will continue to work with our partners to bring scale-out NAS to the front of the market.