Scheduling Precedence Constrained Task Graphs with Non-Negligible Intertask Communication onto Multiprocessors
Issue No. 03 - March (1994 vol. 5)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.277783
<p>The multiprocessor scheduling problem is the problem of scheduling the tasks of aprecedence constrained task graph (representing a parallel program) onto the processorsof a multiprocessor in a way that minimizes the completion time. Since this problem isknown to be NP-hard in the strong sense in all but a few very restricted eases, heuristicalgorithms are being developed which obtain near optimal schedules in a reasonableamount of computation time. We present an efficient heuristic algorithm for schedulingprecedence constrained task graphs with nonnegligible intertask communication ontomultiprocessors taking contention in the communication channels into consideration. Ouralgorithm for obtaining satisfactory suboptimal schedules is based on the classical listscheduling strategy. It simultaneously exploits the schedule-holes generated in theprocessors and in the communication channels during the scheduling process in order toproduce better schedules. We demonstrate the effectiveness of our algorithm bycomparing with two competing heuristic algorithms available in the literature.</p>
Index Termscomputational complexity; multiprocessing systems; multiprocessor interconnectionnetworks; scheduling; heuristic programming; parallel algorithms; graph theory;multiprocessing programs; precedence constrained task graphs; intertask communication;multiprocessors; multiprocessor scheduling problem; parallel program; completion time;heuristic algorithm; list scheduling; communication channels
C. Siva Ram Murthy and C. Selvakumar, "Scheduling Precedence Constrained Task Graphs with Non-Negligible Intertask Communication onto Multiprocessors," in IEEE Transactions on Parallel & Distributed Systems, vol. 5, no. , pp. 328-336, 1994.