Multi-Target vectorization with MTPS c++ generic library

Authors:
Wilfried Kirschenmann;Laurent Plagne;Stéphane Vialle
Affiliations:
SINETICS Department, EDF R&D, France;SINETICS Department, EDF R&D, France;SUPELEC - UMI 2958, France
Venue:
PARA'10 Proceedings of the 10th international conference on Applied Parallel and Scientific Computing - Volume 2
Year:
2010

Citing 5
Cited 1

Arrays in Blitz++

ISCOPE '98 Proceedings of the Second International Symposium on Computing in Object-Oriented Parallel Environments
QUAFF: efficient C++ design for parallel skeletons

Parallel Computing - Algorithmic skeletons
Intel threading building blocks

Intel threading building blocks
Multi-target C++ implementation of parallel skeletons

Proceedings of the 8th workshop on Parallel/High-Performance Object-Oriented Scientific Computing
A Light-weight API for Portable Multicore Programming

PDP '10 Proceedings of the 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing

Optimizing image processing on multi-core CPUs with Intel parallel programming technologies

Multimedia Tools and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.