Title :
Evolutionary algorithms for the synthesis of embedded software
Author :
Zitzler, Eckart ; Teich, Jügen ; Bhattclcharyya, S.S.
Author_Institution :
Comput. Eng. & Networks Lab., Eidgenossische Tech. Hochschule, Zurich, Switzerland
Abstract :
This paper addresses the problem of trading off between the minimization of program and data memory requirements of single-processer Implementations of dataflow programs. Based on the formal model of synchronous dataflow (SDF) graphs, so called single appearance schedules are known to be program-memory optimal. Among these schedules, buffer memory schedules are investigated and explored based on a two-step approach: 1) an evolutionary algorithm (EA) is applied to efficiently explore the (in general) exponential search space of actor firing orders; 2) for each order, the buffer costs are evaluated by applying a dynamic programming post-optimization step (GDPPO). This iterative approach is compared to existing heuristics for buffer memory optimization.
Keywords :
VLSI; buffer storage; data flow graphs; dynamic programming; embedded systems; evolutionary computation; software engineering; storage management; SDF graphs; actor firing orders; buffer costs evaluation; buffer memory optimization; buffer memory schedules; data memory requirements; dataflow programs; dynamic programming post-optimization step; embedded VLSI systems; embedded software synthesis; evolutionary algorithms; exponential search space; iterative approach; program memory requirements; single appearance schedules; single-processer implementations; synchronous dataflow graphs; Costs; Dynamic programming; Dynamic scheduling; Embedded software; Evolutionary computation; Iterative methods; Job shop scheduling; Laboratories; Memory management; Very large scale integration;
Journal_Title :
Very Large Scale Integration (VLSI) Systems, IEEE Transactions on
