Computational Science Technical Note CSTN-159

CSTN Home

Job Parallelism using Graphical Processing Unit individual Multi-Processors and Localised Memory

D. P. Playne and K. A. Hawick

June 2012

Abstract

Graphical Processing Units(GPUs) are usually programmed to provide data-parallel acceleration to a host processor. Modern GPUs typically have an internal multi-processor (MP) structure that can be exploited in an unusual way to offer semi-independent task parallelism providing the MPs can operate within their own localised memory and apply data-parallelism to their own problem subset. We describe a combined simulation and statistical analysis application using component labelling and benchmark it on a range of modern GPU and CPU devices with various numbers of cores. As well as demonstrating a high degree of job parallelism and throughput we find a typical GPU MP outperforms a conventional CPU core.

Keywords: GPU; task parallelism; data parallelism; hybrid parallelism; multi-processor; multi-core.

Full Document Text: PDF version.

Citation Information: BiBTeX database for CSTN Notes.

BiBTeX reference:

@INPROCEEDINGS{CSTN-159,
  author = {D. P. Playne and K. A. Hawick},
  title = {Job Parallelism using Graphical Processing Unit individual Multi-Processors
	and Highly Localised Memory},
  booktitle = {Proc. 19th Int. Conf. on Parallel and Distributed Processing Techniques
	and Applications (PDPTA)},
  year = {2013},
  number = {CSTN-159},
  pages = {PDP2435},
  address = {Las Vegas, USA},
  month = {22-25 July},
  organization = {WorldComp},
  institution = {Computer Science, Massey University},
  keywords = {GPU; task parallelism; data parallelism; hybrid parallelism; multi-processor;
	multi-core}
}


[ CSTN Index | CSTN BiBTeX ]