This Article 
 Bibliographic References 
 Add to: 
Stochastic Bounds on Execution Times of Parallel Programs
October 1991 (vol. 17 no. 10)
pp. 1005-1012

Stochastic bounds are obtained on execution times of parallel programs when the number of processors is unlimited. A parallel program is considered to consist of interdependent tasks with synchronization constraints. These constraints are described by an acyclic directed graph called a task graph. The execution times of tasks are considered to be independently identically distributed (i.i.d.) random variables. The performance measure of interest is the overall execution of the considered parallel program (task graph). Stochastic bound methods are applied to obtain lower and upper bounds on this measure. Another upper bound is obtained for parallel programs having 'new better than used in expectation' (NBUE) random variables as task execution times. NBUE random variables are replaced with exponential random variables of the same mean to derive this upper bound.

[1] V. G. Adlakha and V. G. Kulkarni, "A classified bibliography of research on stochastic Pert networks: 1966-1987," Dep. Operations Res., Univ. North Carolina, Internal Rep.
[2] F. Baccelli and A. Makowski, "Queueing models for systems with synchronization constraints,"Proc. IEEE, vol. 77, pp. 138-162, Jan. 1989.
[3] F. Baccelli and Z. Liu, "On a class of stochastic evolution equations," INRIA, Rep., Mar. 1989.
[4] F. Baccelli and Z. Liu, "On the execution of parallel programs on multiprocessor architectures: a queuing theory approach," INRIA, Rep. 833, Apr. 1988.
[5] F. Baccelli and Z. Liu, "On the stability of a precedence-based queuing discipline," INRIA, Rep. 880, July 1988.
[6] N. Bambos and J. Walrand, "On stability and performance of parallel processing systems,"JACM, vol. 38, no. 2, Apr. 1991.
[7] R. E. Barlow and F. Proschan,Statistical Theory of Reliability and Life Testing. New York: Hold, 1975.
[8] B. Dodin, "Bounding the project completion time distribution in PERT networks,"Oper. Res., vol. 33, no. 4, pp. 862-881, 1985.
[9] G. Fayolle, P. J. B. King, and I. Mitrani, "On the execution of programs by many processors," inProc. Performance'83. Amsterdam: North-Holland, 1983, pp. 217-228.
[10] E. Gelenbe, E. Montagne, R. Suros, and C. M. Woodside, "Performance of block structured parallel programs," inParallel Algorithms and Architectures, M. Cosnardet al., Eds. Amsterdam: North-Holland, 1986, pp. 127-138.
[11] E. Gelenbe, R. Nelson, T. Philips, and A. Tantawi, "The asymptotic processing time for a model of parallel computation," presented at the AFIPS Nat. Comput. Conf., Las Vegas, NV, 1986.
[12] E. Gelenbe and Z. Liu, "Performance analysis of multiprocessors with concurrent tasks," inProc. IFIP WG 7.3 Int. Sem. Performance of Distributed and Parallel Syst., Dec. 1988.
[13] N. Pekergin and J.-M. Vincent, "Calcul de bornes sur le temps d' exécution de graphes de tâches," EHEI, Rep. 89-18, Dec. 1989.
[14] M. S. Ross,Stochastic Processes. New York: Wiley, 1982.
[15] R. A. Sahner and K. S. Trivedi, "Performance and reliability analysis using directed acyclic graphs,"IEEE Trans. Software Eng., vol. SE-13, pp. 1105-1114, Oct. 1987.
[16] D. Stoyan,Comparison Methods of Queues and other Stochastic Models. New York: Wiley, 1984.
[17] J.-M. Vincent, "Stability condition of a service system with precedence constraints between tasks,"Performance Evaluation, vol. 11, Jan. 1991.

Index Terms:
stochastic bound methods; execution times; parallel programs; interdependent tasks; synchronization constraints; acyclic directed graph; task graph; performance measure; NBUE; exponential random variables; directed graphs; parallel programming; performance evaluation; stochastic processes
N. Yazia-Pekergin, J.-M. Vincent, "Stochastic Bounds on Execution Times of Parallel Programs," IEEE Transactions on Software Engineering, vol. 17, no. 10, pp. 1005-1012, Oct. 1991, doi:10.1109/32.99189
Usage of this product signifies your acceptance of the Terms of Use.