New architecture and feature enhancements deliver superior simplicity and flexibility for the management of unstructured file data in enterprise environments
MILPITAS, CA – May 13, 2009 – Gluster today announced the release of GlusterFS Version 2, bringing unprecedented flexibility and value to enterprises facing the challenges of the unstructured file data explosion. Enterprises have deployed GlusterFS in industries such as Healthcare, Energy, Biotech, and Digital Media to cost effectively address their growing data management needs.
GlusterFS runs on industry standard hardware and is backed by a robust subscription services program. Storage building blocks are clustered together, scaling from a few terabytes to multiple petabytes in a single global namespace. GlusterFS is based on a stackable architecture that can be matched to specific application profiles with simple plug-in modules, maximizing application performance across a broad range of deployment scenarios.
“GlusterFS 2.0 exceeded my expectations,” said Dan Parsons, senior systems architect at Synthetic Genomics. “Aside from improving our existing system’s stability and performance, it has allowed us to use our system in scenarios that we didn’t plan for initially, which has saved enormously on hardware costs.”
The modular architecture of GlusterFS is focused on ease of deployment and use, enabling administrators to configure hundreds of terabytes in less than thirty minutes. Unlike typical file systems that have complex kernel dependencies and associated compatibility issues, GlusterFS layers on top of the operating system, simplifying deployment and delivering broad cross-platform compatibility. Growth is managed elegantly by seamlessly adding storage building blocks or disk capacity with built in load balancing for distributing data. Reliability is ensured with features such as on-the-fly self healing and replication to withstand hardware failures.
“The improvements we have made to GlusterFS will get customers in front of the new set of challenges being faced today,” said Anand Babu Periasamy, co-founder and chief technology officer at Gluster. “We recognized the need for greater simplicity in managing unstructured file data in a landscape where applications can have very different requirements for data storage and management.”
About GlusterFS Version 2
This release offers a wide range of new and enhanced features for managing unstructured file data; highlights include:
- No metadata dependencies – metadata management is frequently a source of performance bottlenecks, particularly when there are a very large number of files, and can be a single point of failure in traditional file systems. GlusterFS eliminates this problem with intelligent data placement, an automatic data distribution method that optimizes the use of resources and eliminates metadata bottlenecks.
- Small file optimization – workloads that require management of a large number of very small files present unique performance challenges. GlusterFS addresses this with an integrated high performance database module based on BerkeleyDB that brings the small file management efficiencies of a database into the file system.
- Cloud storage optimizations – enables disk and memory resources in cloud environments to be pooled and managed by GlusterFS. The Non-Uniform File Access (NUFA) feature ensures data is placed within the cloud as close as possible to where it is being accessed.
- Web-embeddable storage – GlusterFS can be tightly integrated with Apache for web services. The included module allows the web server to access filesystem data through native system calls.
Gluster is exclusively focused on simplifying the task of storing and managing the explosion of unstructured data and transforming the economics of storage. GlusterFS is an open source clustered file system that runs on industry standard hardware, delivering multiple times the scalability and performance of conventional storage systems at a fraction of the cost. The GlusterFS architecture is modular and stackable, making it extremely easy to install, customize, and manage. Multiple storage building blocks can be clustered together, supporting petabytes of capacity in a single global namespace. Building a configuration of a few hundred terabytes can be accomplished in less than thirty minutes. For more information please visit [intlink id=”877″ type=”page”]www.gluster.com[/intlink].