January 2007
Return to this issue’s stories

IBM and BlueGene/L Continue Domination of HPCC Competition at SC06

For the second year, IBM and BlueGene/L dominated the High Performance Computing Challenge (HPCC) competition at the annual SuperComputing Conference (SC06). The DOE/NNSA/Lawrence Livermore National Laboratory team, using IBM’s BlueGene/L system, once again swept all four Class 1 (best performance) benchmark awards, proving BlueGene/L’s ability to solve a wide range of computational problems. The four Class 2 productivity awards were split among MIT, IBM, The MathWorks, and the Russian People Friendship University.

The Class 1 HPCC benchmarks—High Performance Linpack, RandomAccess, fast Fourier transform, and STREAM-system—are focused on raw performance. The first place winner of each benchmark was awarded a $500 prize. The HPCC awards were presented at the HPCC award booth at SC06, and the prizes were sponsored by HPCWire magazine. Tom Spelce of Lawrence Livermore accepted the awards on behalf of the team.

Background

In 2003, the DARPA’s High Productivity Computing Systems released the HPCC suite. It examines the performance of HPC architectures using kernels with various memory access patterns of well-known computational kernels. Consequently, HPCC results bound the performance of real applications as a function of memory access characteristics and define performance boundaries of HPC architectures. The suite was intended to augment the TOP500 list, and the results are publicly available for 6 out of 10 of the world’s fastest computers. Implementations exist in most of the major high-end programming languages and environments, accompanied by countless optimization efforts.