This Article 
 Bibliographic References 
 Add to: 
Modeling Speedup (n) Greater than n
April 1990 (vol. 1 no. 2)
pp. 250-256

A simple model of parallel computation which is capable of explaining speedups greater than n on n processors is presented. Necessary and sufficient conditions for these exceptional speedups are derived from the model. Several of the contradictory previous results relating to parallel speedup are resolved by using the model.

[1] G. Amdahl, "Validity of the single processor approach to achieving large scale computing capabilities," inProc. AFIPS Comput. Conf., vol. 30. 1967, pp. 483-485.
[2] J. Barton, "Superlinear performance of the SAXPY kernel on the Silicon Graphics 4D120/GTX," Personal commun., 1988.
[3] R. E. Benner, J. L. Gustafson, and R. E. Montry, "Development and analysis of scientific application programs on a 1024-processor hypercube," Tech. Rep. SAND 88-0317, Sandia National Laboratories, Albuquerque, NM, Feb. 1988.
[4] D. L. Eager, J. Zahorjan, and E. D. Lazowska, "Speedup versus efficiency in parallel systems,"IEEE Trans. Comput., vol. 38, pp. 408-423, Mar. 1989.
[5] H. P. Flatt, "Further comments on a model for parallel processing," Tech. Rep. IBM PASC G320-3503, 1987.
[6] V. Faber, P. M. Lubeck, and A. B. White, Jr., "Superlinear speedup of an efficient sequential algorithm is not possible,"Parallel Comput., vol. 3, pp. 259-260, 1986.
[7] D. H. Grit and J. R. McGraw, "Programming divide and conquer for a MIMD machine,"Software Pract. Exp., vol. 15, no. 1, pp. 41-53, Jan. 1985.
[8] E.F. Gehringer, D.P. Siewiorek, and Z. Segall,Parallel Processing: The Cm*Experience, Digital Press, Billerica, Mass., 1987.
[9] J. L. Gustafson, "Re-evaluating Amdahl's Law,"Commun. ACM, vol. 31, no. 5, pp. 532-533, 1988.
[10] W. A. Kornfeld, "Combinatorially implosive algorithms,"Commun. ACM, vol. 25, no. 10, pp. 734-738, Oct. 1982.
[11] K. Li, "IVY: A shared virtual memory system for parallel computing," inProc. 1988 Int. Conf. Parallel Processing, 1988, pp. 94-101.
[12] J. Lipkis, "Superlinear performance of a monte carlo code on the NYU Ultracomputer," Personal commun., 1988.
[13] T.H. Lai and S. Sahni, "Anomalies in Parallel Branch-and-Bound Algorithms,"Comm. ACM, June 1984, pp. 594-602.
[14] G.-J. Li and B. W. Wah, "Coping with anomalies in parallel branch-and-bound algorithms,"IEEE Trans. Comput., vol. C-35, pp. 568-573, June 1986.
[15] R. Mehrotra and E. F. Gehringer, "Superlinear speedup through randomized algorithms," inProc. 1985 Int. Conf Parallel Processing, 1985, pp. 291-300.
[16] M. Minsky, "Form and content in computer science,"ACM Turing Lect., J. ACM, vol. 17, pp. 197-215, 1970.
[17] E. Miya, "Suggestion on superlinear speed up terminology,"Network News Posting, Dec. 1988.
[18] D. Parkinson, "Parallel efficiency can be greater than unity,"Parallel Comput., vol. 3, pp. 261-262, 1986.
[19] B. R. Preiss and V. C. Hamacher. "Semi-static dataflow," inProc. 1988 Int. Conf. Parallel Processing, 1988, pp. 127-134.
[20] J. Sanguinetti, "Performance of a message-based multiprocessor,"IEEE Comput. Mag., vol. 19, no. 9, pp. 47-55, Sept. 1986.
[21] C. B. Stunkel and W. K. Fuchs, "Analysis of hypercube cache performance using address traces generated by TRAPEDS," inProc. 1989 Int. Conf. Parallel Processing, 1989, pp. 1-33-1-40.
[22] B. J. Smith, "Architecture and applications of the HEP multiprocessor computer," inReal-Time Signal Processing IV, vol. 298. Bellingham, WA: SPIE, 1981, pp. 241-248.
[23] B. W. Weide, "Modeling unusual behavior of parallel algorithms,"IEEE Trans. Comput., vol. C-31, pp. 1126-1130, Nov. 1982.

Index Terms:
Index Termssuperlinearity; superunity; parallel computation; speedups; parallel algorithms; parallel machines; parallel programming; performance evaluation; programming theory
D.P. Helmbold, C.E. McDowell, "Modeling Speedup (n) Greater than n," IEEE Transactions on Parallel and Distributed Systems, vol. 1, no. 2, pp. 250-256, April 1990, doi:10.1109/71.80148
Usage of this product signifies your acceptance of the Terms of Use.