
This Article  
 
Share  
Bibliographic References  
Add to:  
Digg Furl Spurl Blink Simpy Del.icio.us Y!MyWeb  
Search  
 
ASCII Text  x  
V.W. Mak, S.F. Lundstrom, "Predicting Performance of Parallel Computations," IEEE Transactions on Parallel and Distributed Systems, vol. 1, no. 3, pp. 257270, July, 1990.  
BibTex  x  
@article{ 10.1109/71.80155, author = {V.W. Mak and S.F. Lundstrom}, title = {Predicting Performance of Parallel Computations}, journal ={IEEE Transactions on Parallel and Distributed Systems}, volume = {1}, number = {3}, issn = {10459219}, year = {1990}, pages = {257270}, doi = {http://doi.ieeecomputersociety.org/10.1109/71.80155}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, }  
RefWorks Procite/RefMan/Endnote  x  
TY  JOUR JO  IEEE Transactions on Parallel and Distributed Systems TI  Predicting Performance of Parallel Computations IS  3 SN  10459219 SP257 EP270 EPD  257270 A1  V.W. Mak, A1  S.F. Lundstrom, PY  1990 KW  Index Termsperformance prediction; parallel computations; concurrent systems; task system; seriesparallel directed acyclic graph; service centers; queuing network model; simulation; commercial multiprocessor; directed graphs; parallel processing; performance evaluation; queueing theory VL  1 JA  IEEE Transactions on Parallel and Distributed Systems ER   
An accurate and computationally efficient method for predicting the performance of a class of parallel computations running on concurrent systems is described. A parallel computation is modeled as a task system with precedence relationships expressed as a seriesparallel directed acyclic graph. Resources in a concurrent system are modeled as service centers in a queuing network model. Using these two models as inputs, the method outputs predictions of expected execution time of the parallel computation and the concurrent system utilization. The method is validated against both detailed simulation and actual execution on a commercial multiprocessor. Using 100 test cases, the average error of the prediction when compared to simulation statistics is 1.7%, with a standard deviation of 1.5%; the maximum error is about 10%.
[1] P. Heidelberger and K. S. Trivedi, "Queueing network models for parallel processing with asynchronous tasks,"IEEE Trans. Comput., vol. C31, pp. 10991109, Nov. 1982.
[2] P. Heidelberger and K. S. Trivedi, "Analytic queueing models for programs with internal concurrency,"IEEE Trans. Comput., vol. C32, pp. 7382, Jan. 1983.
[3] A. Thomasian and P. Bay, "Queueing network models for parallel processing of task systems," inProc. 1983 Int. Conf. Parallel Processing, H. Siegel and L. Siegel, Eds., IEEE, 1983, pp. 421428.
[4] A. Thomasian and P. Bay, "Analytic queueing network models for parallel processing systems,"IEEE Trans. Comput., vol. C35, no. 12, pp. 10451054, Dec. 1986.
[5] P. Courtois,Decomposability: Queueing and Computer System Applications. New York: Academic, 1977.
[6] J. Mohan, "Performance of parallel programs: Model and analyses," Ph.D. dissertation, Dep. Comput. Sci., CarnegieMellon Univ., July 1984.
[7] C. G. Bell and A. Newell, "The PMS and ISP descriptive systems for computer structures," inAFIPS Conf. Proc. SJCC, 1970, pp. 351374.
[8] M. J. Flynn, "Some computer organizations and their effectiveness,"IEEE Trans. Comput., vol. C21, pp. 948960, Sept. 1972.
[9] C. A. R. Hoare, "Communicating sequential processes,"Commun. ACM, vol. 21, pp. 666677, 1978.
[10] J. L. Baer, "A survey of some theoretical aspects of multiprocessing,"ACM Comput. Surveys, vol. 5, pp. 3180, Mar. 1983.
[11] V. W. Mak, "Performance prediction of concurrent systems," Tech. Rep. CSLTR87344, Comput. Syst. Lab., Stanford Univ., Dec. 1987.
[12] S. Salza and S. S. Lavenberg, "Approximating response time distributions in closed queueing network models of computer performance," inPerformance '81, F. J. Kylstra, Ed. Amsterdam, 1981, pp. 133145.
[13] E. D. Lazawskaet al., Quantitative System PerformanceComputer System Analysis Using Queueing Network Models. Englewood Cliffs, NJ: PrenticeHall, 1984.
[14] M. Reiser and S. Lavenberg, "Mean value analysis of closed multichain queueing networks,"J. ACM, vol. 27, no. 2, Apr. 1980.
[15] K.C. Sevcik and I. Mitrani, "The Distribution of Queueing Network States at Input and Output Instants,"J. ACM, Vol. 28, No. 2, Apr. 1981, pp. 358371.
[16] S. S. Lavenberg and M. Reiser, "Stationary state probabilities of arrival instants for closed queueing networks with multiple types of customers,"J. Appl. Probability, vol. 17, pp. 10481061, Dec. 1980.
[17] J. Zahorjan, "The approximate solution of large queueing network models," Ph.D. dissertation, Univ. Toronto, Aug. 1980.
[18] J. Little, "A proof of the queueing formula L =λW,"Oper. Res., vol. 9, pp. 383387, 1961.
[19] H.D. Schwetman, "CSIM: A CBased, ProcessOriented Simulation Language,"Proc. 1986 Winter Simulation Conf., American Statistical Assoc., Washington, DC, 1986.
[20] P. J. Denning and J. P. Buzen, "The operational analysis of queueing network models,"ACM Comput. Surveys, vol. 10, pp. 225261, Sept. 1978.