November 20, 2008
'Jaguar' demonstrates its power and ease of use.
OAK RIDGE, Tenn., Nov. 20 -- A Cray XT5 supercomputer named Jaguar that runs scientific applications at the U.S. Department of Energy's (DOE's) Oak Ridge National Laboratory (ORNL) placed in three out of four categories at the High-Performance Computing (HPC) Challenge awards, winning two "gold medals" and one "bronze" in this head-to-head competition. Results of the challenge, which measures excellence at handling computing workloads, were announced Nov. 18 in Austin at SC08, an international gathering of supercomputing professionals.
Jaguar won first place for both speed in solving a dense matrix of linear algebra equations (running a software code called High-Performance Linpack, or HPL) and sustainable memory bandwidth?or how many gigabytes per second a node can fetch and store (running the STREAM code). It won third place for speed in executing the Global-Fast Fourier Transformation, a common algorithm used in many scientific applications.
"The Cray Jaguar at ORNL winning two of the HPC Challenge benchmarks shows the power and potential of the computer system for handling some of the most challenging computational science problems," said Jack Dongarra of University of Tennessee-Knoxville and Oak Ridge National Laboratory. "It was able to produce an impressive 902 teraflops [trillion floating point operations per second] on HPL and 330 TB/s [terabytes per second] on STREAMS. Both results leave the second-place IBM Blue Gene/L at Lawrence Livermore National Laboratory far behind and demonstrates the balance between computing and communication bandwidth."
ORNL, along with Cray’s Chapel team led by Brad Chamberlain, shared another award for the most elegant implementation of the HPC Challenge benchmark applications in Cray's Chapel computer language.
John Levesque, director of the Cray Supercomputing Center of Excellence at ORNL, said the HPC Challenge, sponsored by the Defense Advanced Research Projects Agency High Productivity Computing Systems Program, supports hardware and software development needed to effectively use petascale computers, which can execute quadrillions of calculations each second. Similar to the way a decathlon measures performance in ten track and field events, the HPC Challenge measures a computer's ability to excel in execution of a wide variety of components important to running scientific applications.
All of the benchmarks were run in two modes: baseline (no source-code modifications) and optimized (significant source-code modifications). Baselines demonstrate a machine's overall performance and ease of use, whereas optimizations boost performance on one specific aspect of computation. ORNL submitted baselines for Jaguar reflecting the ease of use of the system. In the list of the baseline results, Jaguar ranked as the most powerful machine in three of the four categories and ranked second in the final category, according to the posted results.
"The fact that Jaguar won these awards and placed so highly on the four major benchmarks with the baseline run attests to the superior performance and balance of the system," said Buddy Bland, project director for ORNL's Leadership Computing Facility, which hosts Jaguar. "This is truly a remarkable machine. It is exceptionally powerful in every measure that is important to the scientists who use this machine. Because it is a general-purpose computer that is easy to use, the scientists using this machine have been able to set new performance records on a wide range of science problems in just its first week of availability."
For more information about the HPC Challenge benchmarks, see http://www.hpcchallenge.org.
For more information on Jaguar, see http://www.nccs.gov/jaguar.
Source: Dawn Levy, Oak Ridge National Laboratory
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?