Performance evaluation using data flow graphs for concurrent processing

Document Type

Conference Proceeding

Publication Date

1-1-1992

Abstract

Data Flow Graphs (DFGs) [1,2] are often used for modeling iterative concurrent activities. The acyclic precedence graph (APG) constructed from a DFG can be used to construct an admissible schedule. The iteration period from this schedule is, in general, not optimal. Therefore, it is necessary to find the minimum iteration period (iteration bound) of a schedule [3-6]. This work presents a systematic procedure to find the exact iteration bound for an arbitrary nonterminating DFG. The complexity of the procedure to identify the critical loop and the iteration bound is 0(n3logn), where n is the number of nodes in the DFG. Look-ahead is another technique to enhance concurrency. It inserts delays in the critical loop of the DFG; hence achieving better performance than the iteration bound. To achieve the maximum degree of improvement it is further necessary to identify the next-critical loop. In this paper, a method to identify the next-critical loop of a nonterminating DFG is also presented.

Identifier

84942394640 (Scopus)

ISBN

[0780307208, 9780780307209]

Publication Title

Conference Proceedings IEEE International Conference on Systems Man and Cybernetics

External Full Text Location

https://doi.org/10.1109/ICSMC.1992.271700

ISSN

1062922X

First Page

638

Last Page

643

Volume

1992-January

This document is currently not available here.

Share

COinS