In second place was the former leader, IBM's Sequoia system installed at Lawrence Livermore National Laboratory in the U.S., with a performance of 16.32 petaflops. The third is the Fujitsu K computer set up in the RIKEN Advanced Institute for Computational Science in Japan, while the holder of the fourth position is IBM System Blue Gene/Q named Mira, which is located at Argonne National Laboratory in the U.S., and the fifth is another IBM Blue Gene/Q called Juqueen at the research center Forschungszentrum Jülich in Germany.

On the list of the world's 500 fastest supercomputers are 62 systems that combine graphics and traditional processors, and although AMD's processors are used in the fastest Supercomputer, Intel processors are certainly the most popular, and are used in 76% of the system among the top 500. Maybe this will help AMD to get some additional numbers in sales as the fastest system powered by AMD CPUs is more than a decent promotion.

[Ed – The AMD/nVidia team might have some competition if information about Intel’s Xeon Phi is even close to accurate. The problem will be getting it into these systems, you see nVidia’s Tesla has taken a while to gain the confidence of super computer builders and we imagine that the same will happen with the Phi. Either way, with GPGPUs in the world’s fastest super computer it is only a matter of time before other companies see this as a great upgrade path.]

Tell us what you think in our Forum