About: The acceleration of the sparse matrix-vector multiplication by the region traversal

Facets (new session)
Description
Metadata
Settings
- owl:sameAs
- Inference Rule:

About: The acceleration of the sparse matrix-vector multiplication by the region traversal Goto Sponge Distinct Permalink

An Entity of Type : http://linked.opendata.cz/ontology/domain/vavai/Vysledek, within Data Space : linked.opendata.cz associated with source document(s)

Attributes	Values
rdf:type	skos:Concept http://linked.opendata.cz/ontology/domain/vavai/Vysledek
Description	Tato zpráva popisuje nový přístup ke zrychlení násobení řídké matice vektorem. Princip spočívá ve 3 krocích: 1) rozdělení původní matice na regiony, 2) zvolení efektivního způsobu průchodu těhto regionů, 3) volba optimální způsobu uložení jednotlivých regionů. Naše měření prokázala, že daná metoda dosahuje podstatného zrychlení u matic z různých technických oborů. (cs) Sparse matrix-vector multiplication (shortly spMV) is one of most common subroutines in the numerical linear algebra. The problem is that the memory access patterns during the spMV are irregular and the utilization of cache can suffer from low spatial or temporal locality. This paper introduces new approach for the acceleration the spMV. This approach consists of 3 steps: 1) dividing matrix A into non-empty regions, 2) choosing an efficient way to traverse these regions (in another words choosing an efficient ordering of partial multiplications), 3) choosing the optimal type of storage for each region. In this paper, we describe aspects of these 3 steps in more detail (including fast and time-inexpensive algorithms for all steps). Our measurements proved that our approach gives a significant speedup for almost all matrices arising from various technical areas. Sparse matrix-vector multiplication (shortly spMV) is one of most common subroutines in the numerical linear algebra. The problem is that the memory access patterns during the spMV are irregular and the utilization of cache can suffer from low spatial or temporal locality. This paper introduces new approach for the acceleration the spMV. This approach consists of 3 steps: 1) dividing matrix A into non-empty regions, 2) choosing an efficient way to traverse these regions (in another words choosing an efficient ordering of partial multiplications), 3) choosing the optimal type of storage for each region. In this paper, we describe aspects of these 3 steps in more detail (including fast and time-inexpensive algorithms for all steps). Our measurements proved that our approach gives a significant speedup for almost all matrices arising from various technical areas. (en)
Title	The acceleration of the sparse matrix-vector multiplication by the region traversal The acceleration of the sparse matrix-vector multiplication by the region traversal (en) The acceleration of the sparse matrix-vector multiplication by the region traversal (cs)
skos:prefLabel	The acceleration of the sparse matrix-vector multiplication by the region traversal The acceleration of the sparse matrix-vector multiplication by the region traversal (en) The acceleration of the sparse matrix-vector multiplication by the region traversal (cs)
skos:notation	RIV/68407700:21230/08:03151548!RIV09-MSM-21230___
http://linked.open...avai/riv/aktivita	Z
http://linked.open...avai/riv/aktivity	Z(MSM6840770014)
http://linked.open...iv/cisloPeriodika	4/2008
http://linked.open...vai/riv/dodaniDat	2009
http://linked.open...aciTvurceVysledku	Šimeček, Ivan
http://linked.open.../riv/druhVysledku	J - Článek v odborném periodiku
http://linked.open...iv/duvernostUdaju	S - Úplné a pravdivé údaje nepodléhající ochraně podle zvláštních právních předpisů
http://linked.open...titaPredkladatele	České vysoké učení technické v Praze / Fakulta elektrotechnická
http://linked.open...dnocenehoVysledku	354710
http://linked.open...ai/riv/idVysledku	RIV/68407700:21230/08:03151548
http://linked.open...riv/jazykVysledku	eng - angličtina
http://linked.open.../riv/klicovaSlova	analytical cache model; cache locality; sparse matrix-vector multiplication (en)
http://linked.open.../riv/klicoveSlovo	sparse matrix-vector multiplication analytical cache model cache locality
http://linked.open...odStatuVydavatele	CZ - Česká republika
http://linked.open...ontrolniKodProRIV	[8AFDB0E51B6B]
http://linked.open...i/riv/nazevZdroje	Acta Polytechnica
http://linked.open...in/vavai/riv/obor	IN
http://linked.open...ichTvurcuVysledku	1 (xsd:int)
http://linked.open...cetTvurcuVysledku	1 (xsd:int)
http://linked.open...UplatneniVysledku	2008
http://linked.open...v/svazekPeriodika	48
http://linked.open...iv/tvurceVysledku	Šimeček, Ivan
http://linked.open...n/vavai/riv/zamer	Výzkum perspektivních informačních a komunikačních technologií
issn	1210-2709
number of pages	8 (xsd:int)
http://localhost/t...ganizacniJednotka	21230
is http://linked.open...avai/riv/vysledek of	The acceleration of the sparse matrix-vector multiplication by the region traversal The acceleration of the sparse matrix-vector multiplication by the region traversal The acceleration of the sparse matrix-vector multiplication by the region traversal The acceleration of the sparse matrix-vector multiplication by the region traversal

Faceted Search & Find service v1.16.118 as of Jun 21 2024

Alternative Linked Data Documents: ODE Content Formats:

RDF

ODATA

Microdata

About

OpenLink Virtuoso version 07.20.3240 as of Jun 21 2024, on Linux (x86_64-pc-linux-gnu), Single-Server Edition (126 GB total memory, 112 GB memory in use)
Data on this page belongs to its respective rights holders.
Virtuoso Faceted Browser Copyright © 2009-2024 OpenLink Software