Webinar – InfiniBand-based Storage for Efficient and Scalable High Performance Computing

By Michael Connolly | February 10, 2012 | No comments

I had the recent opportunity on January 25, 2012 to participate with our valuable partner Mellanox Technologies, Ltd. in a joint educational webinar. Mellanox is a leading supplier of high-performance, end-to-end connectivity solutions for data center servers and storage systems...

Posted in: Storage Insights

Ultra-dense storage subsystems – some points to consider

By Doug Donsbach | February 8, 2012 | No comments

In order to minimize the use of space in the modern data center, many storage solution architects and storage subsystem users are turning to innovative packaging schemes to stuff as many hard disk drives (HDDs) and/or solid state drives (SSDs) into as small a volume as...

Posted in: Storage Insights

FDMI™ – File Data Management Interface

By Peter Braam | February 1, 2012 | No comments

Data management is a hugely important aspect of storage, and yet interoperable standard approaches to perform it have been slow in coming.  A few frameworks have become widely used, with DPMI for backup probably leading the pack.  Emerging initiatives such as IRODS hold...

Posted in: Storage Insights

Repairing and Checking File Systems

By Peter Braam | November 28, 2011 | No comments

Checking and repairing large file systems has become one of the biggest challenges facing the technology.   The utility that does this goes by the name of "fsck", and it iterates over the entire content of a file system checking for the integrity of the (relational) data...

Posted in: Storage Insights

Storage Performance: Block vs. File System

By Torben Kling Petersen | November 11, 2011 | No comments

File system performance is often a major component of overall HPC system performance, and is heavily dependent on the nature of the application(s) generating the load. To achieve optimal performance, the underlying file system configuration must be balanced to match the...

Posted in: Storage Insights

Sizing Lustre For Throughput and/or Capacity

By John Fragalla | November 7, 2011 | No comments

Applications will dictate the throughput and capacity of a HPC storage solution using Lustre.  Lustre is a flexible file system that allows for scaling capacity horizontally or vertically to meet the application I/O requirements.  Scaling capacity horizontally, or scaling-out,...

Posted in: Storage Insights

Xyratex Releases Lustre Migrator Tool

By Michael Connolly | November 3, 2011 | No comments

In our continued commitment and support of the Lustre® community, we have announced and released, in mid-October, a Lustre migrator tool that is available for use, licensed under the GPL 2.0, to the entire Lustre community. The Xyratex migrator tool converts Lustre file system...

Posted in: Storage Insights

Reflections from the HPC User Forum – San Diego

By Peter Bojanic | October 31, 2011 | No comments

I've had the privilege of participating in IDC HPC User Forum panel discussions for a few years and I always find the dialog engaging. Last month at the forum meeting in San Diego I joined six other Lustre community leaders from a cross section of organizations to discuss "The...

Posted in: Storage Insights

Why Lustre?

By Torben Kling Petersen | October 25, 2011 | No comments

Lustre is not the only scalable parallel file system in the HPC world but, during it’s 10 year lifespan, it still dominates in eight out of the top ten HPC systems in the world and is in more than 70% of the Top500.  Lustre is open source so it is not tied to a single vendor...

Posted in: Storage Insights

Heading to SBAC-PAD-2001 in Brazil . . .

By Steve Paulhus | October 24, 2011 | No comments

This Thursday, I’ll be speaking at South America’s largest HPC conference about breaking the I/O bottleneck in HPC storage so please join us if you’re planning on to be at the conference. I feel that as an industry, we face many architectural challenges around creating a...

Posted in: Storage Insights

© 2011 Xyratex. All rights reserved.