SKaMPI — Including More Complex Communication Patterns

Michael Haller () and Thomas Worsch ()
Additional contact information
Michael Haller: LIIN, Universität Karlsruhe
Thomas Worsch: LIIN, Universität Karlsruhe

A chapter in High Performance Computing in Science and Engineering ’03, 2003, pp 455-466 from Springer

Abstract: Summary SKaMPI is now an established benchmark for MPI implementations. In autumn 2002 the development of the “new SKaMPI” has started in three major directions: (i) extension of the benchmark to cover more functions of MPI and a redesign of the benchmark allowing it to be extended more easily (thus matching requests from SKaMPI users); (ii) construction of a collection of important algorithm kernels which are not supported by core MPI collective operations; (iii) development of a simulator which (at least partially) supports the simulation of one MPI implementation while using another MPI implementation (possibly running on a different kind of machine). In the present paper we give an overview of the new SKaMPI and describe fast implementations of MPI_Alltoall and of an algorithm for transposing multidimensional matrices as first instances of SKaMPI-Alg.

Keywords: Parallel Machine; Virtual Node; Message Size; Physical Node; Virtual Topology (search for similar items in EconPapers)
Date: 2003
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-642-55876-4_33

Ordering information: This item can be ordered from
http://www.springer.com/9783642558764

DOI: 10.1007/978-3-642-55876-4_33

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().