OpenMP and compilation issue in embedded applications

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Currently embedded systems become more and more important and widely applied to everywhere around us, such as a mobile phone, a PDA, an HDTV, and so on. In this paper, we applied OpenMP to non-traditional benchmarks, i.e. embedded applications in order to examine the applicability of OpenMP in this area. We parallelized embedded benchmarks, called EEMBC, consisting of 5 categories and total 34 applications, and measure their performance in detail. From experiment, we could find 90 parallel sections in 17 applications, but we achieved speedup only in four applications. Since embedded applications consists of a chunk of small loops, we could not get speedup due to large parallelization overheads such as thread management and instruction overheads. Also we show that the OpenMP-inserted parallel code size is much larger than a serial version due to multithreaded libraries, which is critical to embedded systems because of their limited size of memory systems. We discuss an identified critical, but a trivial problem in the current OpenMP specification when we applied OpenMP to these applications.

Original languageEnglish
Pages (from-to)109-121
Number of pages13
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2716
Publication statusPublished - 2003 Dec 1

Fingerprint

Benchmarking
OpenMP
Compilation
Cell Phones
Libraries
Embedded systems
Embedded Systems
Speedup
Benchmark
High definition television
Personal digital assistants
Mobile Phone
Mobile phones
Parallelization
Thread
Performance Measures
Trivial
Specification
Specifications
Data storage equipment

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

@article{51f4a6a752fc4872bdfb14fb5508ff8f,
title = "OpenMP and compilation issue in embedded applications",
abstract = "Currently embedded systems become more and more important and widely applied to everywhere around us, such as a mobile phone, a PDA, an HDTV, and so on. In this paper, we applied OpenMP to non-traditional benchmarks, i.e. embedded applications in order to examine the applicability of OpenMP in this area. We parallelized embedded benchmarks, called EEMBC, consisting of 5 categories and total 34 applications, and measure their performance in detail. From experiment, we could find 90 parallel sections in 17 applications, but we achieved speedup only in four applications. Since embedded applications consists of a chunk of small loops, we could not get speedup due to large parallelization overheads such as thread management and instruction overheads. Also we show that the OpenMP-inserted parallel code size is much larger than a serial version due to multithreaded libraries, which is critical to embedded systems because of their limited size of memory systems. We discuss an identified critical, but a trivial problem in the current OpenMP specification when we applied OpenMP to these applications.",
author = "Jaegeun Oh and Kim, {Seon Wook} and Chulwoo Kim",
year = "2003",
month = "12",
day = "1",
language = "English",
volume = "2716",
pages = "109--121",
journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - OpenMP and compilation issue in embedded applications

AU - Oh, Jaegeun

AU - Kim, Seon Wook

AU - Kim, Chulwoo

PY - 2003/12/1

Y1 - 2003/12/1

N2 - Currently embedded systems become more and more important and widely applied to everywhere around us, such as a mobile phone, a PDA, an HDTV, and so on. In this paper, we applied OpenMP to non-traditional benchmarks, i.e. embedded applications in order to examine the applicability of OpenMP in this area. We parallelized embedded benchmarks, called EEMBC, consisting of 5 categories and total 34 applications, and measure their performance in detail. From experiment, we could find 90 parallel sections in 17 applications, but we achieved speedup only in four applications. Since embedded applications consists of a chunk of small loops, we could not get speedup due to large parallelization overheads such as thread management and instruction overheads. Also we show that the OpenMP-inserted parallel code size is much larger than a serial version due to multithreaded libraries, which is critical to embedded systems because of their limited size of memory systems. We discuss an identified critical, but a trivial problem in the current OpenMP specification when we applied OpenMP to these applications.

AB - Currently embedded systems become more and more important and widely applied to everywhere around us, such as a mobile phone, a PDA, an HDTV, and so on. In this paper, we applied OpenMP to non-traditional benchmarks, i.e. embedded applications in order to examine the applicability of OpenMP in this area. We parallelized embedded benchmarks, called EEMBC, consisting of 5 categories and total 34 applications, and measure their performance in detail. From experiment, we could find 90 parallel sections in 17 applications, but we achieved speedup only in four applications. Since embedded applications consists of a chunk of small loops, we could not get speedup due to large parallelization overheads such as thread management and instruction overheads. Also we show that the OpenMP-inserted parallel code size is much larger than a serial version due to multithreaded libraries, which is critical to embedded systems because of their limited size of memory systems. We discuss an identified critical, but a trivial problem in the current OpenMP specification when we applied OpenMP to these applications.

UR - http://www.scopus.com/inward/record.url?scp=34547357447&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547357447&partnerID=8YFLogxK

M3 - Article

VL - 2716

SP - 109

EP - 121

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SN - 0302-9743

ER -