Multi-target vectorization with MTPS C++ generic library
SINETICS Department, EDF R&D, FRANCE
Applied Parallel and Scientific Computing, Lecture Notes in Computer Science, Volume 7134/2012, 336-346, 2012
@article{kirschenmann2012multi,
title={Multi-target vectorization with MTPS C++ generic library},
author={Kirschenmann, W. and Plagne, L. and Vialle, S.},
journal={Applied Parallel and Scientific Computing},
pages={336–346},
year={2012},
publisher={Springer}
}
This article introduces a C++ template library dedicated at vectorizing algorithms for different target architectures: Multi-Target Parallel Skeleton (MTPS). Skeletons describing the data structures and algorithms are provided and allow MTPS to generate a code with optimized memory access patterns for the choosen architecture. MTPS currently supports x86-64 multicore CPUs and CUDA enabled GPUs. On these architectures, performances close to hardware limits are observed.
March 22, 2012 by hgpu
