HPCwire: NSF Awards SDSC $2.8 Million for Trestles Supercomputer

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

August 19, 2010

NSF Awards SDSC $2.8 Million for Trestles Supercomputer

324-node/100 teraflops system provides a bridge to greater science productivity

Aug. 19 -- The National Science Foundation (NSF) has awarded the San Diego Supercomputer Center (SDSC) at the University of California, San Diego, $2.8 million to build and deploy a new high performance computer system called Trestles.

Designed to increase productivity for a broad spectrum of researchers, Trestles will have 10,368 processor cores, a peak speed of 100 teraflop/s, and 38 terabytes of flash memory. It will begin operations before the end of 2010 and remain in use for three years under the NSF award. The system will work with and span the deployments of SDSC's recently introduced Dash system and its larger Gordon data-intensive system, to become operational in mid-2011.

Like Dash, Trestles will be available to users of the TeraGrid, the nation's largest open-access scientific discovery infrastructure.

"Trestles is appropriately named because it will serve as a bridge between SDSC's unique, data-intensive resources available to TeraGrid users now and into the future," said Mike Norman, SDSC's interim director. "The overarching goal of Trestles is to enable as much productive science as possible in this data-intensive era."

To be configured by SDSC and Appro, Trestles is based on quad-socket, 8-core AMD Magny-Cours compute nodes connected via a QDR InfiniBand fabric. Each of the 324 nodes will have 32 cores, 64 GB of DDR3 memory, and 120 GB of flash memory. With a total of 10,368 cores, Trestles will offer significant compute capability for meeting the science objectives of a large number of users.

In addition, SDSC is taking an innovative approach of complementing the existing capability of TeraGrid systems that favor large users and high system utilization, with a system that has the capacity to deliver science impact for large numbers of TeraGrid users.

"TeraGrid user data show that scientific impact occurs across all allocation levels and that many well-known computational science usage models do not require large-scale jobs or huge allocations," said Richard Moore, deputy director of SDSC and a co-PI on the project.

However, a fair number of computational science approaches require resources with scheduling flexibility and rapid turnaround. "By focusing on core counts of 1,024 or less, Trestles is designed to serve a much larger number of users while simultaneously improving their productivity as measured by turnaround and the number of jobs completed," said Allan Snavely, associate director of SDSC and also a co-PI for the new system.

SDSC's approach with Trestles with regard to its size, allocation range and scheduling practices will also benefit the emerging Science Gateway paradigm for high-performance computing (HPC) system access. Science gateways is a relatively recent phenomenon in supercomputing. Currently led by Nancy Wilkins-Diehr of SDSC, the TeraGrid Gateway program began in 2004 as web portals designed and used by scientists. The program extends the analysis capabilities of these community-designed interfaces through the use of supercomputers, yet insulates users from supercomputing complexities. While gateway usage still represents a small fraction of total TeraGrid usage, about 25 percent of TeraGrid users now access resources via gateways rather than through single-user accounts, according to TeraGrid data.

To ensure that productivity on Trestles remains high, SDSC will adjust allocation policies, queuing structures, user documentation, and training based on a quarterly review of usage and user satisfaction data. SDSC's Dash and Triton Resource clusters use a matrixed pool of expertise in system administration and user support, as well as the SDSC-developed Rocks cluster management software. SDSC will also run the Inca monitoring framework to ensure the system meets all service-level expectations. SDSC's Advanced User Support will establish key benchmarks to accelerate user applications, and subsequently will assist users in tuning and optimizing applications for Trestles.

SDSC announced last November that it won a five-year, $20 million grant from the NSF to build and operate Gordon, the first high-performance supercomputer to employ a vast amount of flash memory to help speed solutions now hamstrung by slower spinning disk technology. Dash, a smaller prototype of Gordon, was deployed in April 2010. Both systems are being integrated by Appro and use a similar design philosophy of combining commodity parts in innovative ways to achieve high-performance architectures.

Trestles is also a world famous surf break in San Diego County and one of the largest remaining undeveloped stretches of Southern California coastline. It is named for the railway bridge that crosses San Mateo Creek, home to a highly diverse and fragile community of plants, birdlife, mammals and reptiles, making it an ideal area to study biodiversity and climate change.

About SDSC

As an Organized Research Unit of UC San Diego, SDSC is a national leader in creating and providing cyberinfrastructure for data-intensive research, and is celebrating its 25th anniversary this year as one of the National Science Foundation's first supercomputer centers. Cyberinfrastructure refers to an accessible and integrated network of computer-based resources and expertise, focused on accelerating scientific inquiry and discovery. SDSC is a founding member of TeraGrid, the nation's largest open-access scientific discovery infrastructure.

Related Links

SDSC: http://www.sdsc.edu/
UC San Diego: http://www.ucsd.edu/
National Science Foundation: http://www.nsf.gov/

-----

Source: Jan Zverina, San Diego Supercomputer Center

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

May 11, 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...