This Article 
 Bibliographic References 
 Add to: 
Probabilistic Analysis of Scheduling Precedence Constrained Parallel Tasks on Multicomputers with Contiguous Processor Allocation
October 2000 (vol. 49 no. 10)
pp. 1021-1030

Abstract—Given a set of precedence constrained parallel tasks with their processor requirements and execution times, the problem of scheduling precedence constrained parallel tasks on multicomputers with contiguous processor allocation is to find a nonpreemptive schedule of the tasks on a multicomputer such that the schedule length is minimized. This scheduling problem is substantially more difficult than other scheduling problems due to precedence constraints among tasks, the inherent difficulty of task scheduling, and processor allocation in multicomputers. We present an approximation algorithm called LLB that schedules tasks level-by-level using the largest-task-first strategy supported by the binary system partitioning scheme to handle the three difficult issues in our scheduling problem. Though algorithm LLB does not have a bounded worst-case performance ratio, we show through probabilistic analysis that LLB has a quite reasonable average-case performance ratio for typical classes of parallel computations. In particular, algorithm LLB has an average-case performance ratio less than two for large scale parallel computations that have wide task graphs (i.e., that exhibit large parallelism).

[1] B.S. Baker, D.J. Brown, and H.P. Katseff, “A 5/4 Algorithm for Two-Dimensional Packing,” J. Algorithms, vol. 2, no. 4, pp. 348-368, 1981.
[2] H.A. David, Order Statistics. New York: John Wiley&Sons, 1970.
[3] A. Feldmann, M.-Y. Kao, J. Sgall, and S.-H. Teng, “Optimal Online Scheduling of Parallel Jobs with Dependencies,” Proc. 25th Ann. ACM Symp. Theory of Computing, pp. 642-651, 1993.
[4] M.R. Garey and D.S. Johnson, Computers and Intractability: A Guide to the Theory of NP-Completeness.New York: W.H. Freeman, 1979.
[5] D.W. Gillies and J.W.-S. Liu, “Greed in Resource Scheduling,” Acta Informatica, vol. 28, pp. 755-775, 1991.
[6] R.L. Graham, “Bounds on Multiprocessing Timing Anomalies,” SIAM J. Applied Math., vol. 2, pp. 416-429, 1969.
[7] K.K. Jain and V. Rajaraman, "Lower and Upper Bounds on Time for Multiprocessor Optimal Schedules," IEEE Trans. Parallel and Distributed Systems, vol. 5, no. 8, pp. 879-886, Aug. 1994.
[8] D.S. Johnson et al., “Worst-Case Performance Bounds for Simple One-Dimensional Packing Algorithms,” SIAM J. Computing, vol. 3, pp. 299-325, 1974.
[9] K. Li, “Stochastic Bounds for Parallel Program Execution Times with Processor Constraints,” IEEE Trans. Computers, vol. 46, no. 5, pp. 630-636, May 1997.
[10] K. Li, “Analysis of the List Scheduling Algorithm for Precedence Constrained Parallel Tasks,” J. Combinatorial Optimization, vol. 3, no. 1, pp. 73-88, 1999.
[11] K. Li, “Analysis of an Approximation Algorithm for Scheduling Independent Parallel Tasks,” Discrete Math. and Theoretical Computer Science, vol. 3, no. 4, pp. 155-166, 1999.
[12] K. Li, “Stochastic Bounds for Parallel Program Execution Times with Processor Constraints,” Proc. Seventh IEEE Symp. Parallel and Distributed Processing, pp. 208-213, Oct. 1995.
[13] K. Li, "Probabilistic Performance Analysis of Scheduling Parallel Tasks with Precedence Constraints on Mesh Connected Multicomputers," Proc. 15th IEEE Int'l Phoenix Conf. Computers and Comm., pp. 358-365, Mar. 1996.
[14] K. Li, “Stochastic Performance Bounds for Scheduling Random Parallel Task Systems on Hypercubes,” Proc. Fourth High-Performance Computing Symp., pp. 199-205, Apr. 1996.
[15] K. Li, “Analyzing the Expected Execution Times of Parallel Programs,” Proc. 12th Ann. ACM Symp. Applied Computing, pp. 488-495, Feb. 1997.
[16] K. Li and Y. Pan, “On Scheduling Precedence Constrained Parallel Tasks on Multiprocessors,” Proc. Int'l Conf. Modelling and Simulation, pp. 63-75, Apr. 1996.
[17] S. Madala and J.B. Sinclair, "Performance of Synchronous Parallel Algorithms with Regular Structures," IEEE Trans. Parallel and Distributed Systems, vol. 2, no. 1, pp. 105-116, Jan. 1991.
[18] J.T. Robinson, “Some Analysis Techniques for Asynchronous Multiprocessor Algorithms,” IEEE Trans. Software Eng., vol. 5, no. 1, pp. 24-31, 1979.
[19] K.S. Trivedi, Probability and Statistics with Reliability, Queuing, and Computer Science Applications. Prentice Hall, 1982.
[20] J. Turek, U. Schwiegelshohn, J. Wolf, and P. Yu,“Scheduling parallel tasks to minimize average response times,”inProc. 5th Annu. ACM-SIAM Symp. Discrete Algorithms, Alexandria, VA, Jan. 1994, pp. 200–209.
[21] J.D. Ullman, “NP-Complete Scheduling Problems,” J. Computer and System Science, vol. 10, pp. 384-393, 1975.
[22] Q. Wang and H. Cheng,“A heuristic of scheduling parallel tasks and its analysis,”SIAM J. Comput., vol. 21, no. 2, pp. 281–294, Apr. 1992.
[23] N. Yazici-Pekergin and J.-M. Vincent, "Stochastic Bounds on Execution Times of Parallel Programs," IEEE Trans. Software Eng., vol. 17, no. 10, pp. 1,005-1,012, Oct. 1991.

Index Terms:
Average-case performance ratio, binary system partitioning, contiguous processor allocation, largest-task-first, parallel task, precedence constraint, probabilistic analysis, task scheduling.
Keqin Li, Yi Pan, "Probabilistic Analysis of Scheduling Precedence Constrained Parallel Tasks on Multicomputers with Contiguous Processor Allocation," IEEE Transactions on Computers, vol. 49, no. 10, pp. 1021-1030, Oct. 2000, doi:10.1109/12.888038
Usage of this product signifies your acceptance of the Terms of Use.