Computing in Science & Engineering, May/June 2011, pp. 92–95
Computing in Science & Engineering magazine aims to meet the need in hard sciences for efficient algorithms, system software, and computer architecture to address large computational problems.
Trends in High-Performance Computing
by Volodymyr Kindratenko and Pedro Trancoso
We can infer the future of high-performance computing (HPC) from the technologies developed today to showcase the leadership-class compute systems — that is, the supercomputers. These machines are usually designed to achieve the highest possible performance in terms of the number of 64-bit floating-point operations per second (flops). Their architecture has evolved from early custom design systems to the current clusters of commodity multisocket, multicore systems. Twice a year, the supercomputing community ranks the systems and produces the Top-500 list, which shows the world's 500 highest performing machines. As we now describe, the technologies used in the top-ranked machines give a good indication of the architecture trends. Read More »