Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
On the performance of parallel matrix factorisation on the hypermesh
Autore:
Al-Ayyoub, A; Ould-Khaoua, M; Day, K;
Indirizzi:
Jordan Univ Sci & Technol, Dept Comp Sci & Informat Syst, Irbid 22110, Jordan Jordan Univ Sci & Technol Irbid Jordan 22110 t Syst, Irbid 22110, Jordan Univ Glasgow, Dept Comp Sci, Glasgow G12 8QQ, Lanark, Scotland Univ Glasgow Glasgow Lanark Scotland G12 8QQ ow G12 8QQ, Lanark, Scotland Sultan Qaboos Univ, Dept Comp Sci, Al Khod, Oman Sultan Qaboos Univ Al Khod Oman boos Univ, Dept Comp Sci, Al Khod, Oman
Titolo Testata:
JOURNAL OF SUPERCOMPUTING
fascicolo: 1, volume: 20, anno: 2001,
pagine: 37 - 53
SICI:
0920-8542(200108)20:1<37:OTPOPM>2.0.ZU;2-D
Fonte:
ISI
Lingua:
ENG
Soggetto:
INTERCONNECTION NETWORKS; DECOMPOSITION; ALGORITHMS; HYPERCUBE; COMPUTER; MESHES;
Keywords:
interconnection networks; multicomputers; orthogonal networks; hypermeshes; matrix factorization; gaussian elimination; performance analysis;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
30
Recensione:
Indirizzi per estratti:
Indirizzo: Al-Ayyoub, A Jordan Univ Sci & Technol, Dept Comp Sci & Informat Syst, POB3030, Irbid 22110, Jordan Jordan Univ Sci & Technol POB 3030 Irbid Jordan22110 Jordan
Citazione:
A. Al-Ayyoub et al., "On the performance of parallel matrix factorisation on the hypermesh", J SUPERCOMP, 20(1), 2001, pp. 37-53

Abstract

Most common multicomputer networks, e.g. d-ary h-cubes, are graph topologies where an edge (channel) interconnects exactly two vertices (nodes). Hypergraphs are a generalisation of the graph model, where a channel interconnects an arbitrary number of nodes. Previous studies have used synthetic workloads (e.g. statistical distributions) to stress the superior performance characteristics of regular multi-dimensional hypergraphs, also known as hypermeshes, over d-ary h-cubes. There has been, however, hardly any study thathas considered real-world parallel applications. This paper contributes towards filling this gap by providing a comparative study of the performance of one of the most common numerical problems, namely matrix factorisation, on the hypermesh, hypercube, and d-ary h-cube. To this end, the paper firstintroduces orthogonal networks as a unified model for describing both the graph and hypergraph topologies. It then develops a generalised parallel algorithm for matrix factorisation and evaluates its performance on the hypermesh, hypercube and d-ary h-cube. The results reveal that the hypermesh supports matrix computation more efficiently, and therefore provides more evidence of the hypermesh as a viable network for future large-scale multicomputers.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 29/03/20 alle ore 01:56:40