Current models for evaluating the relative performance of computer systems are shown to be special cases of a more general formulation. Using this new model, mini- supercomputers and massively parallel machines are compared to selected supercomputer configurations. These results indicate that multiprocessor supercom puters offer performance superior to that of any of the alternatives currently available.
Get full access to this article
View all access options for this article.
References
1.
Amdahl, G.M.1967. Validity of the single-processor approach to achieving large scale computing capabilities. In AFIPS conf. proc, Reston, Va.: AFIPS Press, 30:483-485.
2.
Amdahl, G.M.1988. Limits of expectation. Internat. J. Supercomput. Appl.2(2):88-94.
3.
Clementi, E.1988. Global simulations on scalar, vector, and parallel LCAP-type supercomputers. In Third internat. conf. supercomput. procSt. Petersburg, Fla.: International Supercomputer Institute, 1:63-78.
4.
Denning, P.J.1988. Speeding up parallel processing. Am. Sci.76(4):347-349.
5.
Detrich, J.H., Folsom, D.G., and Rosenzweig, L.J.1988. LCAP/3090 atIBMKingston: evolution of software to support parallel execution. In Third
6.
internat. conf. supercomput. proc. St. Petersburg, Fla.: International Supercomputer Institute, 1:99—108.
7.
Gustafson, J.L., Montry, G.R., and Benner, R.E.1988. Development of parallel methods for a 1024-processor hypercube . SIAM J. Sci. Statist. Comput.9(4):1-32.
Gustafson, J.L.1988b. The scaled-sized model: a revision of Amdahl's law. In Third internat. conf. supercomput. procSt. Petersburg, Fla.: International Supercomputer Institute, 2:130-133.
Logan, D., Saarinen, J., and Clementi, E.1988. LCAP/3090: genesis and evolution of a parallel processing system. In Third internat. conf. supercomput. procSt. Petersburg, Fla.: International Supercomputer Institute , 1:79-98.
