Distributed Prodessing and Dependence Analysis and Class Design for Exploiting Implicit Parallelism in C++


The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 5, No. 6, pp. 1479-1493, Jun. 1998
10.3745/KIPSTE.1998.5.6.1479,   PDF Download:

Abstract

For the parallel execution of C program it is required for the restructuring compiler to collect summary information generated by function calls. It is not so easy to find summary information between called method and extract parallelism for using object reference parameters and returning object reference method. This paper presents an annotation mechanism which helps the parallelization of a sequential C program. GERINFO and SEQ clauses are adopted, respectively, to describe parallel and sequential relations between methods within a C class interfaces. Explicit parallel information being provided, a restructuring compiler then proceeds an efficient program parallelization. Parallel code translator converting C program into parallel code is implemented using the dependence analysis method described in this paper.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
H. D. Young, K. O. Jin, C. Y. Keun, "Distributed Prodessing and Dependence Analysis and Class Design for Exploiting Implicit Parallelism in C++," The Transactions of the Korea Information Processing Society (1994 ~ 2000), vol. 5, no. 6, pp. 1479-1493, 1998. DOI: 10.3745/KIPSTE.1998.5.6.1479.

[ACM Style]
Hwang Deuk Young, Kwon Oh Jin, and Choi Young Keun. 1998. Distributed Prodessing and Dependence Analysis and Class Design for Exploiting Implicit Parallelism in C++. The Transactions of the Korea Information Processing Society (1994 ~ 2000), 5, 6, (1998), 1479-1493. DOI: 10.3745/KIPSTE.1998.5.6.1479.