A heuristic scheduling algorithm of imprecise multiprocessor system with 0/1 constraint

The scheduling problem to satisfy both 0/1 constraint and timing constraints with minimizing the total error is NP-complete when the optional parts have arbitrary processing times. In this paper, we present a heuristic scheduling algorithm of imprecise systems with 0/1 constraint which consist of the communicating tasks running on several processors. The algorithm is based on the program graph which is similar to the one presented in by Natale and Stankovic (1994). To check the schedulability, we apply the Lawler and Moore's theorem. An intensive simulation is done to analyze the performance of suggested algorithm. The results of simulation show that the longest processing first selection strategy outperforms the random or minimal laxity policies.