Perfomance models for blocked sparse matrix-vector multiplication kernels

Karakasis, V; Goumas, G; Koziris, N

dc.contributor.author	Karakasis, V	en
dc.contributor.author	Goumas, G	en
dc.contributor.author	Koziris, N	en
dc.date.accessioned	2014-03-01T02:46:17Z
dc.date.available	2014-03-01T02:46:17Z
dc.date.issued	2009	en
dc.identifier.issn	01903918	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32646
dc.subject	Blocking	en
dc.subject	Performance models	en
dc.subject	Sparse matrix-vector multiplication	en
dc.subject.other	Blocking method	en
dc.subject.other	Blocking performance	en
dc.subject.other	Computational kernels	en
dc.subject.other	High demand	en
dc.subject.other	Indexing structures	en
dc.subject.other	Input matrices	en
dc.subject.other	Memory access	en
dc.subject.other	Memory bandwidths	en
dc.subject.other	Memory subsystems	en
dc.subject.other	Optimization techniques	en
dc.subject.other	Performance Model	en
dc.subject.other	Prefetching	en
dc.subject.other	Shape and size	en
dc.subject.other	Sparse matrices	en
dc.subject.other	Sparse matrix-vector multiplication	en
dc.subject.other	Storage formats	en
dc.subject.other	Knowledge based systems	en
dc.subject.other	Vectors	en
dc.subject.other	Matrix algebra	en
dc.title	Perfomance models for blocked sparse matrix-vector multiplication kernels	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/ICPP.2009.21	en
heal.identifier.secondary	http://dx.doi.org/10.1109/ICPP.2009.21	en
heal.identifier.secondary	5362396	en
heal.publicationDate	2009	en
heal.abstract	Sparse Matrix-Vector multiplication (SpMV) is a very challenging computational kernel, since its performance depends greatly on both the input matrix and the underlying architecture. The main problem of SpMV is its high demands on memory bandwidth, which cannot yet be abudantly offered from modern commodity architectures. One of the most promising optimization techniques for SpMV is blocking, which can reduce the indexing structures for storing a sparse matrix, and therefore alleviate the pressure to the memory subsystem. However, blocking methods can severely degrade performance if not used properly. In this paper, we study and evaluate a number of representative blocking storage formats and present a performance model that can accurately select the most suitable blocking storage format and the corresponding block shape and size for a specific sparse matrix. Our model considers both the memory and computational part of the kernel, which can be non-negligible when applying blocking, and also assumes an overlapping of memory accesses and computations that modern commodity architectures can offer through hardware prefetching mechanisms. © 2009 IEEE.	en
heal.journalName	Proceedings of the International Conference on Parallel Processing	en
dc.identifier.doi	10.1109/ICPP.2009.21	en
dc.identifier.spage	356	en
dc.identifier.epage	364	en