RT info:eu-repo/semantics/conferenceObject
T1 Effortless and Efficient Distributed Data-Partitioning in Linear Algebra
A1 de Blas Cartón, Carlos
A1 González Escribano, Arturo
A1 Llanos Ferraris, Diego Rafael
K1 Informática
K1 Automatic data partition; layouts; distributed systems.
K1 1203 Ciencia de Los Ordenadores
K1 3304 Tecnología de Los Ordenadores
AB This paper introduces a new technique to exploit compositions of different data-layout techniques with Hitmap, a library for hierarchical-tiling and automatic mapping of arrays. We show how Hitmap is used to implement block-cyclic layouts for a parallel LU decomposition algorithm. The paper compares the well-known ScaLAPACK implementation of LU, as well as other carefully optimized MPI versions, with a Hitmap implementation. The comparison is made in terms of both performance and code length. Our results show that the Hitmap version outperforms the ScaLAPACK implementation and is almost as efficient as our best manual MPI implementation. The insertion of this composition technique in the automatic data-layouts of Hitmap allows the programmer to develop parallel programs with both a significant reduction of the development effort and a negligible loss of efficiency.
PB IEEE
SN 978-1-4244-8335-8
YR 2010
FD 2010
LK https://uvadoc.uva.es/handle/10324/75864
UL https://uvadoc.uva.es/handle/10324/75864
LA eng
NO 12th IEEE International Conference on High Performance Computing and Communications, HPCC 2010, 1-3 September 2010, Melbourne, Australia
NO Producción Científica
DS UVaDOC
RD 15-jun-2025