Monday, October 08, 2012

#CHIPS: "Supercomputer Apps Favor MIC over GPU"

Supercomputers used to have massive central-processing units (CPUs) that were lightening quick, but have given way to networked multi-core processors with turned-down clock speeds to save power. Graphic-processing units (GPUs) fit right into this model by virtue of their simple architecture and massive parallelism, but now Intel's many-integrated core (MIC) architecture, used for the first time in the Xeon Phi, promises to do everything that GPUs do, but with x86 software comparibility, prompting a scientist at the National Center for Supercomputing Applications to favor MICs over GPUs: R. Colin Johnson

NCSA’s Lincoln Cluster used Intel Xeon main processors and Nividia Tesla graphic processing units (GPUs).

Here is what Go-Parallel says about MIC versus GPU: While the massively parallel Xeon Phi coprocessor faces supercomputers leveraging Nvidia’s graphic processing units (GPUs), Intel’s many-integrated-core (MIC) architecture will prevail, according to the Senior Research Scientist at the National Center for Supercomputing Applications (NCSA) Innovative Systems Laboratory at the University of Illinois, Urbana-Champaign

In a presentation at the Fifth International Workshop on Parallel Programming Models and Systems Software for High-End Computingn (P2S2) held in Pittsburgh Sept. 10-12, Volodymyr Kindratenko asserted that GPU accelerators will eventually lose out to Intel’s MIC because its architecture only requires a fine-tuning of parallel x86 code already running on supercomputers today...
Further Reading