Parallel Computing of Shared Memory Multiprocessors Based on JOMP

DOI

10.2991/meic-15.2015.346How to use a DOI?

Keywords

parallel computing; JOMP; compiler; speedup; efficiency;

Abstract

Aiming at parallel computing of shared memory multiprocessors based on JOMP, this paper probes the writing, compiling and running of a *.jomp file in detail and describes a complete specification and process for parallel programming. It expounds the format of parallel directive sets, main runtime libraries and their functions, making a further illustration of variable attributes in parallel region. Taking an example of computing values of pi on a lenovo Intel(R) Core(TM) i3-2120, it obtained a 1.73 speedup factor and 86.5% efficiency, and proved the feasibility and effectiveness of this system.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Hong Zhang
AU  - Xiaoming Wang
AU  - Jie Cao
AU  - Changsheng Zhu
PY  - 2015/04
DA  - 2015/04
TI  - Parallel Computing of Shared Memory Multiprocessors Based on JOMP
BT  - Proceedings of the 2015 International Conference on Mechatronics, Electronic, Industrial and Control Engineering
PB  - Atlantis Press
SP  - 1510
EP  - 1514
SN  - 2352-5401
UR  - https://doi.org/10.2991/meic-15.2015.346
DO  - 10.2991/meic-15.2015.346
ID  - Zhang2015/04
ER  -