A novel algorithm of optimal matrix partitioning for parallel dense factorization on heterogeneous processors

Authors:
Alexey Lastovetsky;Ravi Reddy
Affiliations:
School of Computer Science and Informatics, University College Dublin, Belfield, Dublin, Ireland;School of Computer Science and Informatics, University College Dublin, Belfield, Dublin, Ireland
Venue:
PaCT'07 Proceedings of the 9th international conference on Parallel Computing Technologies
Year:
2007

Citing 10
Cited 0

A set of level 3 basic linear algebra subprograms

ACM Transactions on Mathematical Software (TOMS)
ScaLAPACK user's guide

ScaLAPACK user's guide
Heterogeneous distribution of computations solving linear algebra problems on networks of heterogeneous computers

Journal of Parallel and Distributed Computing
A Proposal for a Heterogeneous Cluster ScaLAPACK (Dense Linear Solvers)

IEEE Transactions on Computers
Experiments with mpC: Efficient Solving Regular Problems on Heterogeneous Networks of Computers via Irregulation

IRREGULAR '98 Proceedings of the 5th International Symposium on Solving Irregularly Structured Problems in Parallel
Linear Algebra Algorithms in Heterogeneous Cluster of Personal Computers

HCW '00 Proceedings of the 9th Heterogeneous Computing Workshop
Simulation of Data Distribution Strategies for LU Factorization on Heterogeneous Machines

IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Data Partitioning with a Realistic Performance Model of Networks of Heterogeneous Computers with Task Size Limits

ISPDC '04 Proceedings of the Third International Symposium on Parallel and Distributed Computing/Third International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks
Data Partitioning with a Functional Performance Model of Heterogeneous Processors

International Journal of High Performance Computing Applications
Design and Implementation of the ScaLAPACK LU, QR, and Cholesky Factorization Routines

Scientific Programming

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present a novel algorithm of optimal matrix partitioning for parallel dense matrix factorization on heterogeneous processors based on their constant performance model. We prove the correctness of the algorithm and estimate its complexity. We demonstrate that this algorithm better suits extensions to more complicated, non-constant, performance models of heterogeneous processors than traditional algorithms.