Computer system design using a hierarchical approach to performance evaluation

ABSTRACT

The concept of a hierarchy of performance models is introduced. It is argued that such a hierarchy should consist of models spanning a wide range of accuracy and cost in order to be a cost-effective tool in the design of computer systems. Judicious use of the hierarchy can satisfy the conflicting needs of high accuracy and low cost of performance evaluation. A system design procedure that uses the hierarchy is developed. The concepts developed are illustrated by applying them to a case study of system design. The results of optimizations conducted using a two-level performance model hierarchy and a simple cost model are discussed. In almost all the experiments conducted, the optimization procedure converged to a region very close to a locally optimum system. The efficiency of the procedure is shown to be considerably greater than that of the brute force approach to system design.

REFERENCES

• 1: Anderson, S.F., Earle, J.G., Goldschmidt, R.E., and Powers, D.M. The IBM system 360/91: Floating-point execution unit. IBM J. Res. and Develop. 11 (Jan. 1967), 34-53.
• 2: Anderson, D.W., Sparacio, F.J., and Tomasulo, R.M. The IBM system 360/91: Machine philosophy and instruction handling. IBM J. Res. and Develop. 11 (Jan. 1967), 8-24.
• 3: C. Gordon Bell , A. C. Newell, Computer structures: Readings and examples (McGraw-Hill computer science series), McGraw-Hill Pub. Co., 1971
• 4: Boland, L.T., Granito, G.D., Marcotte, A.V., Messina, B.V., and Smith, J.W. The IBM system 360/91: Storage system. IBM J. Res. and Develop. 11 (Jan. 1967), 54-68.
• 5: Bhandarkar, D.P. A hierarchy of analytical models for complex computer systems. The European Comptng. Conf. Comptr. Syst. Evaluation, Sept. 1976.
• 6: Browne, J.C., et al. Hierarchical techniques for the development of realistic models of complex computer systems. Proc. IEEE, Vol. 63, June 1975, pp. 966-975.
• 7: Courtois, P.J. Decomposability = Queueing and Computer System Applications. Academic Press, N.Y., 1977.
• 8: Dennis, J.B., and Misunas, D.P. A preliminary architecture for a basic data-flow processor. Project MAC Computation Structures Group Memo 102, MIT, Cambridge, Mass., Aug. 1974.
• 9: Draper, N.R., and Smith, H. Applied Regression Analysis. Wiley, N.Y., 1966.
• 10: Gregory, R.T., and Karney, D.L. A Collection of Matrices for Testing Computational Algorithms. Wiley lnterscience, N.Y., 1969.
• 11: Grenander, U., and Tsao, R.F. Quantitative methods for evaluating computer system performance: A review and proposals. In Statistical Computer Performance Evaluation, W. Freiberger, Ed., Academic Press, N.Y., 1972, pp. 3-24.
• 12: IMSL Library 2, Edition 5. Internat. Math. and Statist. Libraries, Inc., 1975.
• 13: Kimbleton, S. A heuristic approach to computer systems performance improvement--a fast performance prediction tool. Proc. 1975 NCC, Vol. 44, AFIPS Press, Arlington, Va., pp. 839-846.
• 14: B. Kumar , E. S. Davidson, Performance evaluation of highly concurrent computers by deterministic simulation, Communications of the ACM, v.21 n.11, p.904-913, Nov. 1978 [doi>10.1145/359642.359646]
• 15: Kumar, B. Performance evaluation of a highly concurrent computer by deterministic simulation. Rep. R-717, Coordinated Sci. Lab., Univ. of Illinois, Feb. 1976.
• 16: Kumar, B. Computer system design using a hierarchical approach to performance evaluation. Rep. R-799, Coordinated Sci. Lab., Univ. of Illinois, Oct. 1977.
• 17: Daniel J. Lasser, Productivity of multiprogrammed computers—progress in developing an analytic prediction method, Communications of the ACM, v.12 n.12, p.678-684, Dec. 1969 [doi>10.1145/363626.363640]
• 18: H. D. Schwetman, Hybrid simulation models of computer systems, Communications of the ACM, v.21 n.9, p.718-723, Sept. 1978 [doi>10.1145/359588.359594]
• 19: Sekino, A. Performance evaluation of multiprogrammed timeshared computer systems. Project MAC Tech. Rep. 103, MIT, Cambridge, Mass., Sept. 1971.
• 20: Svobodova, L. Computer System Performance Measurement and Evaluation Methods: Analysis and Applications. Elsevier, N.Y., 1976.
• 21: Tsao, R.F., Comeau, L.W., and Margolin, B.H. A multi-factor paging experiment. In Statistical Computer Performance Evaluation, W. Freiberger, Ed., Academic Press, N.Y., 1972.
• 22: Tomasulo, R.M. An efficient algorithm for exploiting multiple function units. IBM J. Res. and Develop. 11, 1 (Jan. 1967), 25-33.
• 23: Bernard P. Zeigler, Theory of Modelling and Simulation, Krieger Publishing Co., Inc., Melbourne, FL, 1984

INDEX TERMS

Keywords:
hierarchical modeling, high speed computer systems, optimization algorithms, performance evaluation, system design