Title :
Exploiting Very-Wide Vector Processing for Scientific Applications
Author :
Diavastos, Andreas ; Stylianou, Giannos ; Koutsou, Giannis
Abstract :
Exploiting the recently introduced very-wide vector units of the Xeon Phi coprocessor can potentially increase the scalability for scientific applications. Using lattice QCD compute kernels, the authors find that the performance achieved using the Xeon Phi coprocessors wide vector units is similar to GPGPU performance after appropriate code refactoring, requiring moderate programming effort.
Keywords :
coprocessors; natural sciences computing; quantum chromodynamics; software maintenance; vector processor systems; GPGPU performance; Xeon Phi coprocessor; code refactoring; lattice QCD compute kernels; quantum chromodynamics; scientific applications; very-wide vector processing; Bandwidth; Computer architecture; Coprocessors; Graphics processing units; Scientific computing; Xeon Phi; accelerator; coprocessor; lattice QCD; manycore; scientific computing;
Journal_Title :
Computing in Science Engineering
DOI :
10.1109/MCSE.2015.124
