К содержанию
РецензированоEN2013

MulticoreBSP for C: A High-Performance Library for Shared-Memory Parallel Programming

A. N. Yzelman · R. H. Bisseling · D. Roose

Научные статьи · 2013

Аннотация

The bulk synchronous parallel (BSP) model, as well as parallel programming interfaces based on BSP, classically target distributed-memory parallel architectures. In earlier work, Yzelman and Bisseling designed a MulticoreBSP for Java library specifically for shared-memory architectures. In the present article, we further investigate this concept and introduce the new high-performance MulticoreBSP for C library. Among other features, this library supports nested BSP runs. We show that existing BSP software performs well regardless whether it runs on distributedmemory or shared-memory architectures, and show that applications in MulticoreBSP can attain high-performance results. The paper details implementing the Fast Fourier Transform and the sparse matrix–vector multiplication in BSP, both of which outperform state-of-the-art implementations written in other shared-memory parallel programming interfaces. We furthermore study the applicability of BSP when working on highly non-uniform memory access architectures.

Доказательный фрагмент

«The bulk synchronous parallel (BSP) model, as well as parallel programming interfaces based on BSP, classically target distributed-memory parallel architectures.»

Образец извлечения

Демо-превью

The bulk synchronous parallel (BSP) model, as well as parallel programming interfaces based on BSP, classically target distributed-memory parallel architectures. Полный текст публикации CiteX не хранит — доступ к первоисточнику открывается по DOI согласно лицензии издателя и выбранному плану.

DOI не указан
Цитирований
0
Тип
Журнал
Страна
Узбекистан
Спросить CiteX об этой работе

Смежные источники