Xyratex announces ClusterStor 9000

This market leading efficiency not only provides improved performance but also enables customers to benefit from reduced space, power and administrative costs, up to 60% better than competitive solutions. The ClusterStor 9000 also includes a number of significant Reliability, Availability and Serviceability (RAS) enhancements that improves ease of deployment and system availability to meet the demands of business-critical applications in industries such as Oil and Gas, Financial Services, Genomics, Intelligence, and more. The ClusterStor 9000 will be available through our partners in mid 2014.

Since its release in June 2011, ClusterStor solutions have delivered a new standard in filesystem throughput and storage performance, scalability, and reliability. The innovative and industry proven ClusterStor architecture, which consolidates previously separate server, network components and storage platforms, each with their own discrete software layers, into a tightly integrated modular building block known as a scalable storage unit (SSU), allows customers to linearly scale environment performance non-disruptively from 9 gigabytes per second (GB/s) to over a 1.5 terabytes per second (TB/s) while delivering unmatched efficiency in throughput and cost reduction. This unique architecture is common across the ClusterStor family and is key to alleviating the storage I/O bottleneck, the common choke point in HPC systems.

The ClusterStor 9000 yields over 60GB/s per rack making it 50% faster than the ClusterStor 6000, released in 2012.

"We believe that while performance is always going to a key part of what HPC and Big Data is about, more and more our customers are concerned with the productivity of their infrastructure. We started out to address the very real need of breaking the storage I/O bottleneck, as we identified that legacy systems could simply not meet the performance demands of ever-growing compute clusters. By delivering a truly engineered system that harness the world's leading parallel file system to create a storage solution that is performance optimized, yet easy to deploy, manage and scale, we firmly establish ourselves as the storage performance leaders. Now with the ClusterStor 9000 we are not only dramatically raising the bar once again in terms of performance efficiency and scalability but we are also providing a range of productivity tools that enables greater resiliency to failures at scale both minimizing downtime and improving ease of management which directly translates into improved productivity not only of your storage but also your entire HPC infrastructure investment. Simply put, we are putting the 'Productivity' into HPC Storage", stated Ken Claffey, senior vice president & general manager of the ClusterStor business at Xyratex.

"With the industry's emphasis on scalable systems for Big Data, a lot of companies are coming out with claims of faster I/O performance. What's noteworthy about the Xyratex ClusterStor 9000 announcement is that it goes beyond a dramatic generational improvement in performance and scalability to discuss how that performance can be used effectively to streamline an enterprise's existing work flow", stated Addison Snell, CEO of Intersect 360 Research.

"The acceptance and adoption of ClusterStor since its market introduction is a testament to the experience and stability that comes with a solution provider such as Xyratex", stated Earl Joseph, IDC programme vice president for HPC. "With the release of the CIFS/NFS Gateway, we expect further market acceptance given that the Gateway aims to open avenues to Windows and Linux environments that were previously unavailable for the Lustre file system."

With the addition of the ClusterStor 9000, comes the availability of the following performance features and RAS enhancements to the ClusterStor platform and operating system:

• ClusterStor Grid RAID: this parity declustered RAID feature addition delivers up to 400% faster drive rebuild time to repair times needed to maintain peak performance while supporting the latest high capacity drives.
• ClusterStor Data Integrity Enhancements: these added enhancements of T10 Data Integrity Field (DIF) Type 2 standards provide HPC and Technical Computing users reliable protection of their most valuable asset, their data. Unique to ClusterStor is the ability to include data integrity checking from object storage server HBA to disk drives and hardware assisted cyclic redundancy check (CRC) acceleration providing added resiliency to the file system and improved silent data corruption detection and correction.
• ClusterStor Manager: the new version of this standard ClusterStor feature includes the upgrade of a new dashboard displaying the physical configuration and status health of an entire HPC user's storage system. This data allows for seamlessly integration with other data center management tools providing visibility and management of the ClusterStor solution.

The following new ClusterStor features address the productivity challenges of managing multiple file types and consolidating application work flows with increased efficiency as well as reducing operational and hardware costs:

• ClusterStor Flash Accelerator: this feature transparently streamlines small file and large file IO workloads through application consolidation eliminating the need to separate storage systems thereby reducing Capex and Opex, reduces management cost and complexity, provides linear scalability and includes I/O profiling monitoring and definable policy tools.
• ClusterStor CIFS / NFS Gateway: this gateway feature provides Windows and Linux client access into the ClusterStor Lustre file system and enables heterogeneous file sharing across a single storage pool spanning entry level up to high capacity data storage.
• Lustre Version 2.5 Enhancements: Xyratex integrates available enhancements of this recently released Lustre version into ClusterStor that includes multiple new features such as Hierarchal Storage Management (HSM) allowing data transfer between different storage platform types and facilitating data lifecycle management.