Stephen Withers
Tuesday, 14 July 2009 13:14
Your IT -
Home IT
IBM has shamed its competitors, ranking by far the most environmentally friendly supercomputer maker. Eighteen of the top 20 systems on the Green500 list were made by IBM. The Green500 ranks supercomputers according to the number of floating-point operations per second they can perform for each watt of power consumed.
The bulk of the top 20 spots on the June 2009
Green500 were occupied by IBM BladeCenter and Blue Gene/P systems.
The number one ranking went to a PowerXCell 8i based BladeCenter at the Interdisciplinary Centre for Mathematical and Computational Modeling, University of Warsaw. It was rated at 536.24 MFLOPS/W.
This system is number 422 on the Top500 list of the world's fastest supercomputers.
The two non-IBM systems were the National Astronomical Observatory of Japan's GRAPE-DR accelerator cluster (ranked 5, and using a remarkable two million plus processing elements) and HWW/Universitaet Stuttgart's Xeon X5560 based NEC HPC 140Rb-1 cluster (ranked 20, and the highest-ranked Intel-based system on the list).
It is possible that either of these systems should have been placed higher on the list, as they were rated on peak power consumption rather than measured power.
Compilers of the Green500 list note that while the aggregate power of the 500 systems increased by 15 percent, at the same time the average efficiency increased by 10 percent: "while the supercomputers on the Green500 are collectively consuming more power, they are using the power more efficiently than before."
The world's fastest supercomputer, an IBM system at the Los Alamos National Laboratories in the US, was number four on the Green500.
"Modern supercomputers can no longer focus only on raw performance," said David Turek, vice president, deep computing, IBM.
"To be commercially viable these systems most also be energy efficient. IBM has a rich history of innovation that has significantly increased energy efficiency of our systems at all levels of the system that are designed to simultaneously reduce data center costs and energy use."