CiteULike is a free online bibliography manager. Register and you can start organising your references online.
Tags

Optimization of Collective Reduction Operations Computational Science - ICCS 2004

by: Rolf Rabenseifner

edited by: Marian Bubak, Geert D. Albada, Peter M. A. Sloot, Jack Dongarra

Vol. 3036 (2004), pp. 1-9, doi:10.1007/978-3-540-24685-5_1  Key: citeulike:11299422

Formatted Citation


Show HTML

Likes (beta)

This copy of the article hasn't been liked by anyone yet.

View FullText article


Abstract

A 5-year-profiling in production mode at the University of Stuttgart has shown that more than 40% of the execution time of Message Passing Interface (MPI) routines is spent in the collective communication routines MPI_Allreduce and MPI_Reduce. Although MPI implementations are now available for about 10 years and all vendors are committed to this Message Passing Interface standard, the vendors’ and publicly available reduction algorithms could be accelerated with new algorithms by a factor between 3 (IBM, sum) and 100 (Cray T3E, maxloc) for long vectors. This paper presents five algorithms optimized for different choices of vector size and number of processes. The focus is on bandwidth dominated protocols for power-of-two and non-power-of-two number of processes, optimizing the load balance in communication and computation.


FemkeDeBackere's tags for this article

Citations (CiTO)

No CiTO relationships defined

X There are no reviews yet

X Posting History


X Export records

Privacy Statement | Terms & Conditions
CiteULike organises scholarly (or academic) papers or literature and provides bibliographic (which means it makes bibliographies) for universities and higher education establishments. It helps undergraduates and postgraduates. People studying for PhDs or in postdoctoral (postdoc) positions. The service is similar in scope to EndNote or RefWorks or any other reference manager like BibTeX, but it is a social bookmarking service for scientists and humanities researchers.