Dense linear algebra kernels on heterogeneous platforms: redistribution issues

Authors:
Olivier Beaumont;Arnaud Legrand;Fabrice Rastello;Yves Robert
Affiliations:
LIP, UMR CNRS-ENS Lyon-INRIA 5668, École Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France;LIP, UMR CNRS-ENS Lyon-INRIA 5668, École Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France;LIP, UMR CNRS-ENS Lyon-INRIA 5668, École Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France;LIP, UMR CNRS-ENS Lyon-INRIA 5668, École Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France
Venue:
Parallel Computing - Parallel matrix algorithms and applications
Year:
2002

Citing 6
Cited 5

Introduction to parallel computing: design and analysis of algorithms

Introduction to parallel computing: design and analysis of algorithms
A high-performance matrix-multiplication algorithm on a distributed-memory parallel computer, using overlapped communication

IBM Journal of Research and Development
Tiling a rectangle with the fewest squares

Journal of Combinatorial Theory Series A
ScaLAPACK user's guide

ScaLAPACK user's guide
Computers and Intractability; A Guide to the Theory of NP-Completeness

Computers and Intractability; A Guide to the Theory of NP-Completeness
Matrix-Matrix Multiplication on Heterogeneous Platforms

ICPP '00 Proceedings of the Proceedings of the 2000 International Conference on Parallel Processing

Fast solution of large N × N matrix equations in an MIMD-SIMD hybrid system

Parallel Computing - Special issue: Parallel and distributed scientific and engineering computing
Self-adapting software for numerical linear algebra and LAPACK for clusters

Parallel Computing - Special issue: Parallel and distributed scientific and engineering computing
Adaptive data parallel computing on workstation clusters

Journal of Parallel and Distributed Computing
Heuristics for work distribution of a homogeneous parallel dynamic programming scheme on heterogeneous systems

Parallel Computing - Heterogeneous computing
ABCLib_DRSSED: A parallel eigensolver with an auto-tuning facility

Parallel Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Redistribution algorithms for dense linear algebra kernels on heterogeneous platforms are considered. In this context, processor speeds may well vary during the execution of a large kernel, which requires efficient strategies for redistributing the data along the computations. The proposed strategy is to redistribute data after some well-identified static phases and therefore is neither fully static nor fully dynamic. An optimal algorithm (under some assumptions) for redistributing data when computing the product of two matrices is presented.