Abstract
A compile-time technique is outlined that yields low-cost, analytic performance models, intended for crude scalability analysis and first-order system design. The approach extends current static techniques by accounting for any type of resource contention that may occur. In this paper we report on the accuracy of the prediction method in terms of theory, simulation experiments, as well as measurements on a distributed-memory machine. It is shown that for series-parallel computations with random resource access patterns, the average prediction error is limited well within 50 % regardless the system parameters, where traditional compile-time methods yield errors up to orders of magnitude.
Preview
Unable to display preview. Download preview PDF.
References
V.S. Adve, Analyzing the Behavior and Performance of Parallel Programs. PhD thesis, University of Wisconsin, Madison, Dec. 1993. Tech. Rep. #1201.
M. Ajmone Marsan, G. Balbo and G. Conte, “A class of Generalized Stochastic Petri Nets for the performance analysis of multiprocessor systems,” ACM Tr. on Comp. Syst., 2, May 1984, pp. 93–122.
M. Annaratone, C. Pommerell and R. Rühl, “Interprocessor communication and performance in distributed-memory parallel processors,” in Proc. 16th Symp. on Comp. Archit, May 1989, pp. 315–324.
D. Atapattu and D. Gannon, “Building analytical models into an interactive prediction tool,” in Proc. Supercomputing '89, 1989, pp. 521–530.
V. Balasundaram, G. Fox, K. Kennedy and U. Kremer, “A static performance estimator to guide data partioning decisions,” in Proc. 3rd ACM SIGPLAN Symp. on PPoPP, Apr. 1991.
T. Fahringer and H.P. Zima, “A static parameter-based performance prediction tool for parallel programs,” in Proc. 7th ACM ICS, Tokyo, July 1993, pp. 207–219.
E. Gelenbe, E. Montagne, R. Suros and C.M. Woodside, “Performance of block-structured parallel programs,” in Parallel Algorithms and Architectures, North-Holland, 1986, pp. 127–138.
A.J.C. van Gemund, “Performance prediction of parallel processing systems: The Pamela methodology,” in Proc. 7th ACM ICS, Tokyo, July 1993, pp. 318–327.
A.J.C. van Gemund, “Compiling performance models from parallel programs,” in Proc. 8th ACM ICS, Manchester, July 1994, pp. 303–312.
A.J.C. van Gemund, “On the accuracy of compile-time performance prediction,” Tech. Rep. 1-68340-44(1994)02, Delft University of Technology, Sept. 1994.
N. Götz, U. Herzog and M. Rettelbach, “Multiprocessor and distributed system design: The integration of functional specification and performance analysis using stochastic process algebras,” LNCS, 729, Springer, 1993.
F. Hartleb and V. Mertsiotakis, “Bounds for the mean runtime of parallel programs,” in Proc. 6th Int. Conf. Modelling Techniques and Tools for Comp. Perf. Eval., Edinburgh, Sept. 1992, pp. 197–210.
R.W. Hockney, “Performance parameters and benchmarking of supercomputers,” Parallel Computing, 17, 1991, pp. 1111–1130.
H. Jonkers, A.J.C. van Gemund and G.L. Reijns, “A probabilistic approach to parallel system performance modelling,” in Proc. 28th HICSS, 1995, pp. 412–421.
W. Kreutzer, System simulation, programming styles and languages. Addison-Wesley, 1986.
H.X. Lin and H.J. Sips, “Parallel direct solution of large sparse systems in finite element computations,” in Proc. 7th ACM ICS, Tokyo, July 1993, pp. 261–270.
V.W. Mak and S.F. Lundstrom, “Predicting performance of parallel computations,” IEEE Trans. on PDS, 1, July 1990, pp. 257–270.
A.D. Maloney, V. Mertsiotakis and A. Quick, “Automatic scalability analysis of parallel programs based on modeling techniques,” LNCS, 794, Springer, 1994, pp. 139–158.
C.L. Mendes, J-C. Wang and D.A. Reed, “Automatic performance prediction and scalability analysis for data parallel programs,” in Proc. 2nd. Workshop on Autom. Data Layout and Perf. Pred., Houston, Apr. 1995.
R.A. Sahner and K.S. Trivedi, “SPADE: A tool for performance and reliability evaluation,” in Modelling Techn. and Tools for Perf. Anal. 1986, pp. 147–163.
V. Sarkar, Partitioning and Scheduling Parallel Programs for Multiprocessors. Pitman, 1989.
F. Sötz, “A method for performance prediction of parallel programs,” LNCS, 457, Springer, 1990, pp. 98–107.
H. Wabnig and G. Haring, “Petri net performance models of parallel systems — methodology and case study,” LNCS, 817, Springer, 1994, pp. 301–312.
K-Y. Wang, “A framework for static, precise performance prediction for superscalar-based parallel computers,” in Proc. 4th Int. Workshop on Compilers for Par. Comput., Delft, Dec. 1993, pp. 413–427.
J. Zahorjan et al., “Balanced job bound analysis of queueing networks,” CACM, 25, Feb. 1982, pp. 134–141.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
van Gemund, A.J.C. (1995). Compile-time performance prediction of parallel systems. In: Beilner, H., Bause, F. (eds) Quantitative Evaluation of Computing and Communication Systems. TOOLS 1995. Lecture Notes in Computer Science, vol 977. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0024323
Download citation
DOI: https://doi.org/10.1007/BFb0024323
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60300-9
Online ISBN: 978-3-540-44789-4
eBook Packages: Springer Book Archive