OpenMP parallel programming using dual-core embedded system

  • Kyung Min Lee
  • , Tae Houn Song
  • , Seung Hyun Yoon
  • , Key Ho Kwon
  • , Jae Wook Jeon

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Multi-cores have recently been applied to smartphones, as well as PCs; the performance of electronic devices is improving. However, there is no improvement in performance, if the existing sequential program is executed on multi-core processors. Parallel programming is necessary to use multi-core processors well. Already some companies are doing projects to parallel their own programs that are executed in the PC environment, for example, Adobe Systems Inc., Autodesk Inc., and Epic Games Inc. However, very few embedded field studies, focus on parallel programming. In this paper, we study a parallel programming model, OpenMP, and parallel programs that can be benchmarked to multi-core processors of embedded boards using OpenMP. We execute parallel programs on a dual-core embedded system. We analyze the performance of sequential programs and parallel programs by SERPOP analysis. Finally, we reduce the execution time of programs by a mean of 111%.

Original languageEnglish
Title of host publicationICCAS 2011 - 2011 11th International Conference on Control, Automation and Systems
Pages762-766
Number of pages5
StatePublished - 2011
Event2011 11th International Conference on Control, Automation and Systems, ICCAS 2011 - Gyeonggi-do, Korea, Republic of
Duration: 26 Oct 201129 Oct 2011

Publication series

NameInternational Conference on Control, Automation and Systems
ISSN (Print)1598-7833

Conference

Conference2011 11th International Conference on Control, Automation and Systems, ICCAS 2011
Country/TerritoryKorea, Republic of
CityGyeonggi-do
Period26/10/1129/10/11

Keywords

  • Benchmark
  • Embedded System
  • Geometric Mean
  • Multi-Core
  • OpenMP
  • Parallel Programming
  • SERPOP

Fingerprint

Dive into the research topics of 'OpenMP parallel programming using dual-core embedded system'. Together they form a unique fingerprint.

Cite this