Change search
ReferencesLink to record
Permanent link

Direct link
A matrix-type for performance-portability: STATE OF THE ART IN SCIENTIFIC COMPUTING
KTH, School of Information and Communication Technology (ICT), Microelectronics and Information Technology, IMIT.
2006 (English)In: APPLIED PARALLEL COMPUTING: STATE OF THE ART IN SCIENTIFIC COMPUTING / [ed] Dongarra, J; Madsen, K; Wasniewski, J, BERLIN: SPRINGER-VERLAG BERLIN , 2006, Vol. 3732, 237-246 p.Conference paper (Refereed)
Abstract [en]

When matrix computations are expressed in conventional programming languages, matrices are almost exclusively represented by arrays, but arrays are also used to represent many other kinds of entities, such as grids, lists, hash tables, etc. The responsibility for achieving efficient matrix computations is usually seen as resting on compilers, which in turn apply loop restructuring and reordering transformations to adapt programs and program fragments to target different architectures. Unfortunately, compilers are often unable to restructure conventional algorithms for matrix computations into their block or block-recursive counterparts, which are required to obtain acceptable levels of performance on most current (and future) hardware systems. We present a datatype which is dedicated to the representation of dense matrices. In contrast to arrays, for which index-based element-reference is the basic (primitive) operation, the primitive operations of our specialized matrix-type are composition and decomposition of/into submatrices. Decomposition of a matrix into submatrices (of unspecified sizes) is a key operation in the development of block algorithms for matrix computations, and through direct and explicit expression of (ambiguous) decompositions of matrices into submatrices, block algorithms can be expressed explicitly and at the same time the task of finding good decomposition parameters (i.e., block sizes) for each specific target system, is exposed to and made suitable for compilers.

Place, publisher, year, edition, pages
BERLIN: SPRINGER-VERLAG BERLIN , 2006. Vol. 3732, 237-246 p.
, Lecture Notes in Computer Science, ISSN 0302-9743 ; 3732
Keyword [en]
linear algebra subprograms, data-flow analysis, array, transformations, optimizations, algorithms, set
National Category
Computer Science
URN: urn:nbn:se:kth:diva-42113DOI: 10.1007/11558958_28ISI: 000237003200028ScopusID: 2-s2.0-33745328109ISBN: 3-540-29067-2OAI: diva2:446811
7th International Workshop on State of the Art in Scientific Computing. Lyngby, DENMARK. JUN 20-23, 2004
QC 20111010Available from: 2011-10-10 Created: 2011-10-05 Last updated: 2011-10-10Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Drakenberg, N. Peter
By organisation
Microelectronics and Information Technology, IMIT
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 7 hits
ReferencesLink to record
Permanent link

Direct link