HPCwire: NLR Helps Enable SC09 Bandwidth Challenge Winners

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

December 11, 2009

NLR Helps Enable SC09 Bandwidth Challenge Winners

Provided majority of bandwidth for research exhibits at leading, global high-performance computing conference

CYPRESS, Calif., Dec. 9 -- National LambdaRail (NLR), the coast-to-coast, high-performance network owned by the research and education community, provided ultra-high speed, wide-area network infrastructure and services for winners in all categories of the prestigious Bandwidth Challenge awards at the Supercomputing 2009 (SC09) conference, the world's premier conference for high-performance computing, networking, storage and analysis.

NLR also provided most of the bandwidth powering the research demonstrations at the event. Of a total of approx. 400 Gigabits per second (Gbps) to the SC09 exhibit hall in Portland, Ore., NLR provided 210 Gbps for 13 research organizations collaborating with more than 3 dozen additional research teams from 12 countries.

"We congratulate the University of Illinois at Chicago, Caltech and partners, and the University of Tokyo on winning this distinguished award," said NLR President and CEO Glenn Ricart. "NLR is very pleased to have helped support all winning teams with its advanced networking."

University of Illinois at Chicago: Winner in Overall and Manifold-Process Implementations Categories

The University of Illinois at Chicago's National Center for Data Mining (NCDM), together with the International Center for Advanced Internet Research (iCAIR) and the Naval Research Laboratory, prevailed in both the overall and the "rich, manifold-process implementations including diverse mechanisms" categories. The partners demonstrated three applications to show efficient bandwidth utilization in distributed, data-intensive applications. The first demo processed very large datasets over 256 servers in 4 data centers at more than 100 Gbps using the open source software Sector/Sphere and UDT, developed by NCDM.

The second demo was a cloud-based image rendering application delivering very high-resolution visualization (computed by remote cloud systems) over long-distance Infiniband and IPv6. A hardware implementation of UDT was deployed to support the long-distance Infiniband protocol. The third demo showcased a lightweight UDT variant called UDX, which can transfer data at 9.x Gbps using a single connection over a 10 Gbps network with 200 millisecond (ms) Round Trip Time (RTT). Overall, the team achieved 25 Gbps sustained throughput over a 200 ms RTT, 12,000 mile path utilizing only seven servers on the SC09 floor.

"NLR is a long-standing, close partner for us, providing the 10-Gigabit Ethernet (GE) circuits for the Open Cloud Testbed which was the platform for our SC09 demo and which continues to be the testbed for our cloud standards work," said Robert Grossman, director of the Laboratory for Advanced Computing at the University of Illinois at Chicago and Managing Partner of the Open Data Group.

The University of Illinois team conducted its SC09 demonstration on the Open Cloud Testbed of the Open Cloud Consortium, which runs on 10-GE circuits from NLR with nodes in Washington, D.C., Chicago, San Diego and Los Angeles. For SC09, NLR extended the Open Cloud Testbed, the world's only 10-GE WAN cloud, to the show floor in Portland.

For additional details on NCDM and its partners' accomplishments, visit http://www.ncdm.uic.edu/.

Caltech and Partners: Massive Data Transfers Category Winner

Caltech and its partners won in the massive data transfers category of the Bandwidth Challenge with a record-breaking demonstration of storage-to-storage data transfer over wide-area networks. The high-energy physics (HEP) team achieved a bi-directional peak throughput of 119 Gbps and a data flow of more than 110 Gbps that could be sustained indefinitely among clusters of servers on the show floor and at Caltech, Michigan, San Diego, Florida, Fermilab, Brookhaven, CERN, Brazil, Korea, and Estonia, demonstrating how long-range networks can be used efficiently to support leading-edge science on four continents. NLR provided 6 dedicated and 5 shared 10-GE circuits, out of Caltech's total of 15 10-GE circuits, between Portland, Chicago, Los Angeles and Seattle.

"NLR played a pivotal role ensuring that we had the bandwidth we needed to support our leading-edge data transfer goals, paving the way for next-generation networks and the next round of scientific discoveries at the LHC in the coming years." said Harvey Newman, Caltech professor of physics, head of the HEP team and co-lead of U.S. Large Hadron Collider network (US LHCNet), and chair of the US LHC Users Organization.

Following the Bandwidth Challenge the team continued its tests and demonstrated a world-record data transfer between the Northern and Southern hemispheres, sustaining 8.26 Gbps on each of two 10 Gbps links linking Sâo Paulo and Miami.

The following institutions collaborated with Caltech: the University of Michigan, Fermilab, Brookhaven National Laboratory, CERN, San Diego (UCSD), Florida (UF and FIU), Brazil (Rio de Janeiro State University, UERJ, and State Universities of Sâo Paulo, USP), Korea (Kyungpook National University, KISTI), Estonia (NICPB) and the National University of Science and Technology in Pakistan (NUST).

For further information on the results from Caltech and its partners, visit http://supercomputing.caltech.edu/.

University of Tokyo: Impact on Target Communities Category Winner

The University of Tokyo was presented with the Bandwidth Challenge award for strongest impact on target communities. The team demonstrated data transfer using HTTP between Tokyo and Portland with the UsadaFox system installed on an ordinary PC running an HTTP server-client system based on Apache and Firefox, and attained a 6.5 Gbps transfer speed, which is 1000 times faster compared to the typical Apache server and Firefox browser combination. This successful demonstration has confirmed that UsadaFox enables ordinary people with ordinary PC environments to enjoy 10 Gbps Internet, which in the past was only an option for researchers with high-performance computers. Also, the University of Tokyo demonstrated that UsadaFox is able to use a Web browser running an ordinary, single TCP/IP stream, and does not need special environments such as a Grid system or parallel file system.

"NLR's platform and responsiveness to our particular requirements exceeded our expectations. The loss-less quality of NLR networks and its management was essential to win the award," said Professor Kei Hiraki, head of the University of Tokyo's Data Reservoir Project.

For the University of Tokyo, NLR provided 10-GE, Layer 2 circuits between Sunnyvale and Portland, with the Japanese research and development testbed network, JGN2plus, handling the international network segments from Japan.

Further information on the University of Tokyo's results and award: http://data-reservoir.adm.s.u-tokyo.ac.jp/press-20091127/index-en.html.

Award Criteria

Competitors in the Bandwidth Challenge Award were evaluated on the following attributes:

Degree of integration (between files systems, partner sites, distributed resources).
Raw data transfer performance (best 15 minute window based on 30 second samples with 5 minutes sliding average).
Effective use of resources (based on stated objectives).
Technological reach / degree of innovation.
Robustness / transferability.
Applicability / usability (to other fields, particularly outside computing and hard sciences).

About National LambdaRail (NLR)

Owned by the U.S. research and education community and dedicated to serving the needs of researchers and research groups, NLR is the innovation platform for a wide range of academic disciplines and public-private partnerships. NLR's coast-to-coast, high-performance network infrastructure offers unrestricted usage and bandwidth, a choice of cutting-edge network services and applications, and customized support for individual researchers and projects. For more information, visit www.nlr.net.

-----

Source: National LambdaRail

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...