IBM is in 17 of the top 20 slots, and 65 of the top 100, with Cell, iDataPlex, and Blue Gene. The number 1 system on the list, an IBM at Forschungszentrum Juelich (FZJ), is ranked with 773.38 MFLOPS/W, and shows up at #5 on the Top500 list. For comparison, the number one on the Top500 (ORNL’s Cray Jaguar system) is 56 on the Green500 with only 253 MFLOPS/W. Which just goes to show that it’s all about what you measure.
As the Green500 web site points out, 8 of the top 10 systems are accelerator-based (6 use Cell, and 2 use NVIDIA GPUs).
Green500 co-founder Wu Feng, associate professor of computer science and electrical & computer engineering at the College of Engineering at Virginia Tech explained the significance of the ‘fuel efficiency’ of these accelerator-based supercomputers. “The accelerator-based supercomputers on The Green500 List produce an average efficiency of 554 MFLOPS/W whereas the other measured supercomputers on the list produce an average efficiency of 181 MFLOPS/W. That makes the accelerator-based supercomputers on the Green500 more than three times more energy efficient than their non-accelerated counterparts on the list” Feng said.
The most efficient x86-only cluster in the world is from my alma mater, Mississippi State University. Go dawgs!