UC San Diego physics professor Frank Wuerthwein with Gordon

Gordon, the supercomputer launched last year by the San Diego Supercomputer Center (SDSC) at the University of California, San Diego, recently completed its most data-intensive task so far – rapidly processing raw data from almost one billion particle collisions as part of a project to help define the future research agenda for the Large Hadron Collider (LHC).

Under a partnership between a team of UC San Diego physicists and the Open Science Grid (OSG), a multi-disciplinary research partnership funded by the US Department of Energy and the National Science Foundation, Gordon has been providing auxiliary computing capacity by processing massive data sets generated by the Compact Muon Solenoid (CMS), one of two large general-purpose particle detectors at the LHC used by researchers to find the elusive Higgs particle.

This exciting project has been the single most data-intensive exercise yet for Gordon since we completed large-scale acceptance testing back in early 2012,’ said SDSC director Michael Norman, who is also an astrophysicist involved in research studying the origins of the universe. ‘I’m pleased that we were able to make Gordon’s capabilities available under this partnership between UC San Diego, the OSG and the CMS project.”

This story appears here as part of a cross-publishing agreement with Scientific Computing World.

In this video from Moabcon 2013, Adaptive Computing CEO Robert A. Clyde presents the first annual Adaptie Awards.

The Adaptie Awards provide a platform to recognize our customers for their cutting-edge achievements in the HPC and cloud communities,” said Rob Clyde, CEO of Adaptive Computing. “The myriad applications of Moab is truly inspiring, and we are proud to count such esteemed individuals and organizations as our customers.”

Adaptie Winners:

• Best use of Moab in HPC: National Oceanic and Atmospheric Administration (NOAA).
• Best use of Moab in the Cloud: Bank of America.
• Lifetime Achievement Award: Don Maxwell, ORNL.

The HPC systems team lead at ORNL was honored for his contributions to the HPC industry. Maxwell was instrumental in providing both requirements and testing for the initial port of Moab to the Cray X-series platform. He was awarded the ACM Gordon Bell Prize in 2008 as a member of a team led by Thomas Schulthess that used the ORNL Jaguar supercomputer to achieve 400+ TFlop/s sustained performance. Jaguar ran the Moab HPC Suite, and now, Maxwell uses the Moab HPC Suite on Titan, ORNL’s latest supercomputer, which was declared the world’s fastest in the latest semiannual TOP500 list, securing the No. 1 position.

Read the Full Story or check out the Moabcon 2013 Video Gallery.

Over at RusBase, Ben Hopkins writes that Russian supercomputer vendor T-Platforms has been added to the US Department of Commerce’s “list of organizations and individuals acting contrary to the national security or foreign policy interests of the United States.”

In practice this means that, in accordance with the Export Administration Regulations on March 8 for “T-Platforms’ presumption introduced a ban on the export and re-export of processor capacity over 5 GFlops, as well as any goods and products manufactured in the U.S. or to U.S. technology in other countries. For export or re-export “T-Platforms” are required to obtain a license to the Bureau of Industry and Security.

T-Platforms is one of the top supercomputer vendors in Russia, and their “Lomonosov” supercomputer at Moscow State University was ranked #26 on the TOP500 in November 2012. Last October, the company delivered its first U.S. system at New York State University.

Read the Full Story.

Asetek has announced that the US Department of Energy’s National Renewable Energy Laboratory (NREL) will install Asetek’s RackCDU direct-to-chip ‘hot water’ liquid-cooling system as a retrofit to NREL’s Skynet HPC cluster.

As part of this liquid-cooling retrofit, the cluster will be relocated into the new data centre at the Energy Systems Integration Facility (ESIF) in Golden, Colorado, which is designed to be the most energy efficient data centre in the world, with a PUE (power usage effectiveness) of 1.06.

The data centre at ESIF will use ‘warm water’ (75 degrees Farenheit) liquid cooling to operate servers and to recover waste heat for use as the primary heat source for the building office space and laboratories. The higher liquid temperatures used by Asetek’s RackCDU (105oF) will improve waste-heat recovery and reduce water consumption for the data centre.

Because of RackCDU’s design, these performance improvements will be achieved without the need for a customised server design. The system will be installed as a drop-in retrofit to existing air-cooled servers and racks.

RackCDU is a hot water, direct-to-chip, data centre liquid cooling system that enables cooling energy savings of up to 80 per cent and density increases of 2.5x when compared to modern air cooled data centres. RackCDU removes heat from CPUs, GPUs, memory modules and other hot spots within servers and takes it out of the data centre using liquid where it can be cooled for free using outside air, or recycled to generate building heat and hot-water.

By retrofitting an existing air-cooled HPC cluster with RackCDU, NREL will reduce the cooling energy required to operate this system, reduce water usage in the cooling system and increase the server density within the cluster, reducing floor-space and rack infrastructure requirements.

Ambient water temperature in the hydronic system is a critical factor in data centre efficiency and sustainability,’ said Steve Hammond, director of the Computational Science Center at NREL. Starting with warmer water on the inlet side can create an opportunity for enhanced waste-heat recovery and reduced water consumption, and in many locations can be accomplished without the need for active chilling or evaporative cooling, which could lead to dramatically reduced cooling costs.”

This story appears here as part of a cross-publishing agreement with Scientific Computing World.

With March Madness behind us, April is shaping up to be quite the month for key HPC events. At insideHPC, we’re all set to provide you with news, interviews, and videos from four great shows in the next four weeks.

• Moabcon 2013 (April 8-11). Set in beautiful Park City, Utah, Moabcon is Adaptive Computing’s annual user conference. Featuring keynotes and in-depth technical discussions, we’ll be bringing you video presentations from NCSA, NERSC, NICS, LBNL and more. The event runs April 8-11.

• Lustre User Group (April 16-18). Set in downtown San Diego, LUG 2013 is the annual user group meeting sponsored by Open SFS and EOFS. The event will feature a Lustre Advanced User seminar and two days of presentations on select Lustre features, upcoming enhancements, site-specific experiences using Lustre, and more. We’ll bring you as many video presentations from the event as we can.
  • Pre-LUG 2013 Party Announcement. Traveling to LUG? Intel and Aeon Computing are co-sponsoring a Pre-LUG party that first Monday night, Apr 15th. Beer and Lustre, a great combination! We’ll be in the Omni Hotel, the same location as LUG itself, up on the 10th floor, catching up with old Lustre friends and hopefully make some new ones. Please join us! For more information, email [email protected]

• OpenFabrics Alliance International Developer Workshop (April 21-24). Set in Monterey, this workshop is open to anyone with the desire to provide collaborative input into the direction and mechanics of OpenFabrics Software.

• HPC User Forum (April 29- May 1). Set in Tucson, the HPC User Forum has a mission promote the health of the global HPC industry and address issues of common concern to users. The organization has grown to 150 members. It is directed by a volunteer Steering Committee of HPC users from government, industry and academia, and is operated for the users by market analyst firm IDC.

We’ll see you out on the road!