The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing
November 02, 2007
RENO, Nev., Oct. 31 -- Obsidian, the leader in InfiniBand range-extension, is providing Oak Ridge National Labs (ORNL) with a means of testing long-haul InfiniBand over UltraScience NET (USN). Dr. Nageswara Rao, USN's co-Principal Investigator, said, "We are interested in technologies that will help us move increasingly large datasets between Department of Energy (DOE) labs and facilities. The Longbows on USN are posting remarkable performance numbers and offer a very attractive alternative to TCP/IP."
The DOE operates many of the world's largest "capability" supercomputers, producing and consuming data files that are growing exponentially in size as ever larger machines are deployed -- increased data capacities enable larger, higher-fidelity physics simulations. However, this growth trend presents a critical problem to the DOE: "When supercomputer users need to move super-sized files, shared-infrastructure TCP/IP-based networks are unable to handle the bandwidth overload," remarked Dr. Rao. "Clustering techniques for processors and storage have allowed the computing machinery to scale, but the conventional approaches to networking them together have not kept pace."
ORNL has been evaluating Obsidian Longbow XR devices since September 2006, running a variety of tests over USN, which is well suited as a next-generation networks testbed. Comprising a pair of OC-192 SONET channels between Oak Ridge, Atlanta, Chicago, Seattle and Sunnyvale, it can be configured to provide loops of varying lengths. Results have been highly encouraging, according to Dr. Rao: "Longbow XRs configure trivially easily, simply appearing as two-port InfiniBand switches that completely hide the optical Wide Area Network (WAN) from the InfiniBand servers and storage. Using InfiniBand's Remote Direct Memory Access (RDMA), we found we could stream data across all 8,600 miles of USN with an essentially flat bandwidth-distance curve, right out of the box. To see this kind of plug-and-play performance with no in-situ tuning, and independent of distance, is in striking contrast to normal network performance."
Dr. David Southwell, President and CEO of Obsidian Research, said, "We are most pleased with Dr. Rao's results using Longbow XRs across USN, and look forward to working with the DOE in the future." Work is already underway to integrate the InfiniBand-enabled Lustre file system to provide file level native InfiniBand transport between DOE locations at very high speeds.
Dr. Rao is providing live demonstrations of this capability across USN between ORNL in Tennessee and Reno during SC07 in the Oak Ridge National Labs booth.
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. For more information, see http://www.obsidianresearch.com/.
About UltraScience Net
The next generation DOE large-scale science applications have requirements that will drive extreme networking, in terms of sheer high throughputs as well as dynamically stable connections. The UltraScience Net is (USN) an experimental research testbed funded by DOE Office of Science to enable the development of hybrid optical networking and associated technologies to meet the unprecedented demands of large-scale science applications. For more information, see http://www.csm.ornl.gov/ultranet/.
-----
Source: Obsidian Research Corp.
(Digg, Technorati, more)
PGI Accelerator™ Fortran 95/03 and C99 compilers for x64+NVIDIA
Accelerate applications on x64+GPU platforms by adding OpenMP-like compiler directives to existing Fortran and C programs. Available now for Linux, MacOS and Windows. Download a free 15 day trial.
Platform HPC Workgroup Manager
Platform HPC Workgroup Manager integrates all the cluster productivity tools you need to deploy, run and manage your HPC environment.
C-DAC announces plans for a petaflop system; IBM researchers are working on vertical integration techniques to extend Moore's Law another 15 years. We recap those stories and more in our weekly wrapup.
Read More...
The Moscow State University supercomputer, Lomonosov, has been selected for a high-performance makeover, with the goal of tripling its processing power to achieve petaflop-level performance in 2010. T-Platforms, who developed and manufactured the supercomputer, is the odds-on favorite to lead the project.
Read More...
Right on schedule, Intel has launched its Xeon 5600 processors, codenamed "Westmere EP." The 5600 represents the 32nm sequel to the Xeon 5500 (Nehalem EP) for dual-socket servers. Intel is touting better performance and energy efficiency, along with new security features, as the big selling points of the new Xeons.
Read More...
Mar 19 | OfficialWire | New super to support intelligence work Down Under. Read more...
Mar 18 | ChannelWeb | Westmere parts already showing up in HPC machines. Read more...
Mar 17 | The Register | But what about the tier ones? Read more...
Mar 17 | Cadalyst Magazine | A new generation of workstations is changing the nature of technical computing. Read more...
Mar 17 | Linux Magazine | Latest iteration of Sun Grid Engine able to tap into Cloud. Read more...
Jan 12 | | In-depth look at vSMP Foundation server virtualization technology, technical implementation, use cases and capabilities. The technical whitepaper provides an architectural overview and details on the three vSMP Foundation products: vSMP Foundation for SMP, vSMP Foundation for Cluster and vSMP Foundation for Cloud.
Jan 18 | | This white paper discusses Gore’s copper cable assemblies, and how they continue to exceed the standards for providing reliable, cost-effective solutions for high-performance computer applications.
Join this online panel discussion for live Q&A with leading industry experts, analysts, and end-users to discuss the latest innovations, best practices, barriers to implementation, and measurable benefits of server virtualization with a particular focus on today's real world solutions.
Learn about scalable fault-tolerant architectures and examples of energy efficient and scalable supercomputing clusters using dual QDR InfiniBand to combine capacity computing with network failover capabilities with the help of programming languages such as MPI and a robust Linux cluster management package.
LIVE@SCO9: The IBM team discusses new innovations in hardware, software and services that help clients better understand their workloads and get insight from their R&D efforts. Technology demonstrations include the soon-to-be-released Power7 HPC processor, the DCS990 system with 2.4 petabytes of storage, the xCAT management tool, secure HPC cloud computing and more. Winners of two HPCwire Readers' and Editors’ Choice Awards! Take the IBM virtual tour at SC09 or more information go online to: http://www-03.ibm.com/systems/deepcomputing/sc09.html