Structuring distributed relation-based computations with SCDRC

Botta, N.; Ionescu, C.; Linstead, C.; Klein, R.

Item

ITEM ACTIONSEXPORT

Add to Basket

Local TagsRelease HistoryDetailsSummary

Released

Report

Structuring distributed relation-based computations with SCDRC

Authors

Botta, N.
Potsdam Institute for Climate Impact Research and Cooperation Partners;

Ionescu, C.
Potsdam Institute for Climate Impact Research and Cooperation Partners;

Linstead, C.
Potsdam Institute for Climate Impact Research and Cooperation Partners;

Klein, R.
Potsdam Institute for Climate Impact Research and Cooperation Partners;

External Resource

https://www.pik-potsdam.de/en/output/publications/pikreports
(Publisher version)

Fulltext (restricted access)

There are currently no full texts shared for your IP range.

Fulltext (public)

There are no public fulltexts stored in PuRe

Supplementary Material (public)

There is no public supplementary material available

Citation

Botta, N., Ionescu, C., Linstead, C., Klein, R. (2006): Structuring distributed relation-based computations with SCDRC, (PIK Report ; 103), Potsdam : Potsdam-Institut für Klimafolgenforschung, 61 p.

Cite as: https://publications.pik-potsdam.de/pubman/item/item_13823

Abstract

In this report we present a set of software components for distributed relation-based computations (SCDRC). We explain how SCDRC can be used to structure parallel computations in a single-program multiple-data computational environment.

First, we introduce relation-based algorithms and relation-based computations as generic patterns in scientific computing. We then discuss the problems that have to be solved to parallelize such patterns and propose a high-level formalism for specifying these problems.

This formalism is then applied to derive parallel distributed relation-based computations. These are implemented in the C++ library SCDRC. We present language independent elements of SCDRC and discuss C++ specific aspects of its design and architecture.

Finally, we discuss how to use SCDRC in a simple application and provide preliminary performance figures.