Abstract
Optimization of vector-intensive applications for the CRAY X-MP/Y-MP often requires arranging the operations to take full advantage of such architectural features as the memory system, independent memory ports, chaining, and independent functional units. Estimation of performance is not straightforward since many operations can occur concurrently. As a tool for making trades between vector algorithms, a method has been developed and used successfully at E-Systems Inc. to predict the execution time of a sequence of vector operations without resorting to actual code development. This method reduced our software development time, produced significantly more efficient code, and provided for a systematic approach to optimization. The performance estimation is generally accurate to within 10% and accounts for memory conflicts that result from fixed stride references.
Similar content being viewed by others
References
Bucher, I.Y., and Simmons, M.L. 1991. Measurement of memory access contentions in multiple vector processor systems. In Proc., Supercomputing '91 Conf. (Albuquerque, N. Mex., Nov. 18–22), IEEE Comp. Soc. Press, pp. 806–817.
Calahan, D.A. 1988. An analysis of vector startup access delays. IEEE Trans. Comps., 37 (Sept.): 1134–1137.
Cheng, H. 1989. Vector pipelining, chaining, and speed on the IBM 3090 and CRAY X-MP. IEEE Comp. Mag. (Sept.): 31–42, 44, 46.
Cheung, T., and Smith, J.E. 1986. A simulation study of the CRAY X-MP memory system. IEEE Trans. Comps., 35 (July): 613–622.
Cray Research, Inc. 1982. CRAY X-MP Series Mainframe Reference Manual, HR-0032.
Cray Research, Inc. 1988. CRAY Y-MP System Programmer Reference Manual, CSM-0400-000.
Fatoohi, R.A. 1989. Vector performance analysis of three supercomputers: CRAY-2, CRAY Y-MP, and ETA10-Q. In Proc., Supercomputing '89 Conf. (Reno, Nev., Nov. 13–17), ACM Press, pp. 779–788.
Gatland, I.R. 1986. A “vector calculator” training tool. In Supercomputers: Algorithms, Architectures, and Scientific Computation (T. Tajima and F.A. Matsen, eds.), Univ. of Tex. Press, Austin, Tex., pp. 143–150.
Hainline, A.R. 1991. Optimization techniques on CRI supercomputers. In Conf. Proc., Cray User's Group, Spring '91 (London, Apr. 22–26), pp. 320–323.
Thompson, S.R. 1991. Vector performance calculation as a tool for optimization. In Conf. Proc., Cray User's Group, Spring '91 (London, Apr. 22–26), pp. 315–319.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hainline, A.R., Thompson, S.R. & Halcomb, L.L. Vector performance estimation for CRAY X-MP/Y-MP supercomputers. J Supercomput 6, 49–70 (1992). https://doi.org/10.1007/BF00128642
Issue Date:
DOI: https://doi.org/10.1007/BF00128642