HPCwire: Ames Laboratory Demonstrates Ultra-Fast PVFS Transport

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

Leading HPC
Solution Providers

HPCwire >> Off the Wire

November 02, 2007

Ames Laboratory Demonstrates Ultra-Fast PVFS Transport

RENO, Nev., Nov. 1 -- Obsidian, the leader in InfiniBand range-extension, is pleased to announce very high performance from the Longbow XR InfiniBand router devices deployed at the Scalable Computing Laboratory at Department of Energy (DOE)'s Ames Laboratory in Iowa. Researcher Troy Benjegerdes is demonstrating reliable bulk data transport between remote InfiniBand subnets using the open-source Parallel Virtual File System (PVFS). PVFS was developed by the DOE to harness the parallel storage architectures behind cluster-based supercomputers.

"Like many other government labs and agencies, the DOE continues to face massive growth in the volumes of data it must store and distribute", said Troy Benjegerdes, "and the traditional use of TCP/IP over Ethernet is proving to be very difficult to scale to long-haul, high-bandwidth connections. As the DOE moves into the Petascale Computing Era (10^15+ calculations per second) it is clear that a more efficient and scalable transport solution is needed."

"Obsidian developed the InfiniBand range-extending Longbow XR to transparently provide native InfiniBand connections across optical Wide Area Networks (WANs)," said Dr. David Southwell, President and CEO of Obsidian Research, "InfiniBand flow control has proven to be highly robust and efficient across high-latency 10 Gbits/s long-haul networks -- yielding much higher efficiencies than TCP/IP while preserving InfiniBand's low CPU overhead semantics, such as Remote Direct Memory Access (RDMA)."

Troy Benjegerdes adapted PVFS to support the Wide Area RDMA services provided by Obsidian's Longbow XR products. In the Ames Laboratory's booth in Reno, the link can be seen in action over BOREAS and Internet2 WAN circuits reaching back to Ames Laboratory's high performance test bed PVFS storage system in Ames, Iowa. The Longbow XRs are operating in router mode, such that each end of the WAN link can preserve local InfiniBand subnets, simplifying network security and management.

About Obsidian

Obsidian Research Corporation and the Obsidian Longbow LP are the developers of Longbow, a series of InfiniBand range extension products. Longbow technology allows an InfiniBand fabric, normally a short-range network used in high-performance computing, to be extended via optical fiber over varying distances. Longbow connects across Campus, Metro or Global networks to offer unparalleled high-bandwidth, low-latency access to InfiniBand compute and storage resources. Obsidian is available online at http://www.obsidianresearch.com.

About Scalable Computing Laboratory

The SCL was created in 1989 as a joint effort of the DOE through Ames Laboratory and Iowa State University through the Center for Physical and Computational Mathematics. SCL's mission is to improve parallel computing through clustering techniques for use in scientific and engineering computation. For more information, visit http://www.scl.ameslab.gov/.

-----

Source: Obsidian Research Corporation

Article Tools

Share & Save Options

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCWire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCWire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Feature Articles

Complete Genomics Takes Off

Last week, San-Francisco-based Complete Genomics came out of stealth mode to become the first provider of large-scale human genome sequencing services. HPCwire recently asked company representatives a few questions about their new offering.
Read More...

Intel Grabs NetEffect Assets, Becomes iWARP Player

Intel has acquired the assets of NetEffect, an Austin-based company that makes iWARP-capable adapters. Intel will inherit NetEffect's product portfolio, which includes 1 and 10 GbE accelerated adapters, 10 GbE adapters for blade configurations as well as a 10 GbE ASIC.
Read More...

Woven Launches New 10 GbE Switch

Woven Systems has added a new 10 Gigabit Ethernet top-of-rack switch to its product lineup. The TRX 200 is aimed at high performance datacenter environments requiring a scalable Ethernet fabric.
Read More...

Read more HPCwire features...

Top Headlines

Hazy Computing

Oct 15 | Linux Magazine | Today machines manage what we cannot. Are we dependent upon results or processes we do not understand? Read more...

Reaching For the Exa-Scale

Oct 15 | International Science Grid This Week | Exa-scale computing is probably years away. But GPUs and volunteer grids may provide a shortcut. Read more...

New Visualization Laboratory Debuts on UT Austin's Main Campus

Oct 14 | Texas Advanced Computing Center | TACC has unveiled a new visualization laboratory capable of reproducing terascale data sets with exceptional clarity and resolution. Read more...

High-Performance Nonsense

Oct 13 | Computerworld | Microsoft will have to overcome Windows' historical baggage if its new HPC Server 2008 offering is to be acceptable to users. Read more...

ORNL's Breakthroughs in Cray Machines Make it Hard to Beat

Oct 13 | Knoxville News Sentinel | Oak Ridge National Laboratory has petaflop computing in sight as it upgrades its 'Jaguar' supercomputer. Read more...

Read more headlines...

Featured Whitepapers

Panasas® Tiered Parity™ Architecture

Sep 04 | | Disk drives are approximately 250 times denser today than a decade ago. This is good news for users who are creating, manipulating and storing more data than ever before. It gives them an opportunity to derive more value from their stored data and lowers the capital acquisition and operating expense associated with that data.

SUSE® Linux Enterprise Server for High Performance Computing

Sep 05 | | The excellent scalability features of Linux, in addition to robust security and performance makes it an excellent choice for server systems, especially in the high performance computing area.

View the White Paper Library

Multimedia

Video White Paper: Architecting a Better Network Storage Solution

BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.