Abstract
The performance of a multiprocessor system greatly depends on the bandwidth of its memory architecture. In this paper, uniform memory architectures with various interconnection networks including crossbar, multiple-buses and generalized shuffle networks are studied. We propose a general method based on the Markov chain model by assuming that the blocked memory requests will be redistributed to the memory modules in the next memory cycle. This assumption results in an analysis with lower complexity where the number of states is linearly proportional to the number of processors. Moreover, it can provide excellent estimation on the system power and memory bandwidth for all three types of interconnection networks as compared with the simulation results in which the blocked memory requests are resubmitted to the same memory module. Comparisons also show that our method is more general and precise than most existing analysis methods. The method is further extended to estimate the performance of multiprocessor system with caches. The approximation results are also shown to be remarkably good.
Similar content being viewed by others
References
M. Dubois and S. Thakkar,Cache architectures in tightly coupled multiprocessors, IEEE Computer, June 1990, pp. 9–11.
L. N. Bhuyan and D. P. Agrawal,Design and performance of generalized interconnection networks, IEEE Transaction on Computer, vol. C-32, Dec. 1983, pp. 1081–1090.
D. H. Lawire,Access and alignment of data in an array processor, IEEE Transaction on Computer, vol. C-24, Dec. 1975, pp. 1145–1155.
H.J. Siegel,Analysis techniques for multiprocessors, IEEE Transaction on Computer, vol. C-29, Oct. 1980, pp. 771–780.
D. P. Bhandarkar,Analysis of memory interference in multiprocessor, IEEE Transaction on Computer, vol. C-24, Sept. 1975, pp. 897–908.
W. D. Strecker,Analysis of the instruction execution rate in certain computer structures, Ph.D. dissertation, Carnegie-Mellon University, 1970.
D. Yen, J. Patel and E. Davidson,Memory interference in synchronous multiprocessor systems, IEEE Transaction on Computer, vol. C-31, Nov. 1982, pp. 1116–1121.
J. H. Patel,Performance of processor-memory interconnections for multiprocessors, IEEE Transaction on Computer, vol. C-30, Oct. 1981, pp. 771–780.
T. Lang, M. Valero and I. Alegre,Bandwidth of crossbar and multiple-bus connection of multiprocessors, IEEE Transaction on Computer, vol. C-31, Dec. 1982, pp. 1227–1234.
L. N. Bhuyan,An analysis of processor-memory interconnection networks, IEEE Transaction on Computer, vol. C-34, Mar. 1985, pp. 279–283.
C. R. Das and L. N. Bhuyan,Bandwidth availability of multiple-bus multiprocessors, IEEE Transaction on Computer, vol. C-34, Oct. 1985, pp. 918–926.
J. H. Patel,Analysis of multiprocessors with private cache memories, IEEE Transaction on Computer, vol. C-31, April 1982, pp. 296–304.
T. N. Mudge, J. P. Hayes, G. D. Buzzard and D. C. Winsor,Analysis of multiple-bus interconnection networks, Proceedings of IEEE 1984 International Conference on Parallel Processing, Aug. 1984, pp. 228–232.
T. N. Mudge and H. B. Al-Sadoun,A Semi-Markov model for the performance of multiple-bus systems, IEEE Transaction on Computer, vol. C-34, Oct. 1985, pp. 934–942.
J. P. Sheu and W. T. Chen,Performance analysis of multiple bus interconnection networks with hierarchical requesting model, Proceeding of 8th IEEE Conference on Distributed Computing Systems, June 1988, pp. 138–144.
Q. Yang and S. G. Zaky,Communication performance in multiple-bus system, IEEE Transaction on Computer, vol. C-37, July 1988, pp. 848–853.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Chen, JJ., Wang, CS. & Chou, CR. A general performance analysis method for uniform memory architectures. BIT 33, 536–560 (1993). https://doi.org/10.1007/BF01990534
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01990534