Performance of Cross-Bar Multiprocessor Systems

Siomalas,; Bowen,

doi:10.1109/tc.1983.1676301

Cited by 22 publications

(9 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…= 1 and Cs = 1 / 2, which is the coefficent of variation for the service time Sof the common memory: PE = 45.37 % and QPE = 35.80 % were obtained. This yields L = 0.91 % for a 1 % simulation error with 95 % confidence, and by equation (13), we obtain Ns -24,000.…”

Section: Simulation Experimentsmentioning

confidence: 73%

Simulation experiments for performance analysis of multiple-bus multiprocessor systems with nonexponential service times

Önyüksel

Irani

1989

SIMULATION

View full text Add to dashboard Cite

A simulation model (program) is constructed for performance analysis of multiple-bus multiprocessor systems with shared memories. It is assumed that the service time of the common memory is either hypo-or hyperexponentially distributed. Processing efficiency is used as the performance index. To investigate the effects of different service time distributions on the system performance, comparative results are obtained for a large set of input parameters. The simulation results show that the error in approximating the memory access time by an exponentially distributed random variable is less than 6% if the coefficient of variation is less than 1, but it increases drastically with this factor if it is greater than 1.

show abstract

Section: Simulation Experimentsmentioning

confidence: 73%

Simulation experiments for performance analysis of multiple-bus multiprocessor systems with nonexponential service times

Önyüksel

Irani

1989

SIMULATION

View full text Add to dashboard Cite

show abstract

“…This leads to the use of a discrete Markov Chain approach [1]- [3] for modeling it. In the second policy, the rejected request is simply discarded [4]- [6]. Clearly, this strategy makes the modeling easy and usually suggests probabilistic approaches.…”

Section: Introductionmentioning

confidence: 99%

A comprehensive performance evaluation of crossbar networks

Youn

Chen²

1993

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

Interconnection network is one of the major factors which determine the performance of a multiprocessor system. Therefore, it is imperative to have a model by which the efficiency of the interconnection network can be correctly evaluated. In this paper, we present a comprehensive model for evaluating crossbar networks, where the memory bandwidth and processor acceptance probability are primary measures considered. The proposed analytical model includes all important network control policies such as the bus arbitration and rejected request handling policies as well as home memory concept. Computer simulation validates the correctness. Our study also confirms that the home memory and dynamic bus arbitration policy improve the network performance.Zndex Terms-Analytical model, bus arbitration, crossbar network, home memory, interconnection network, memory bandwidth, processor acceptance probability, rejected request policy.

show abstract

“…Some of these studies have resulted in simple expressions for the throughput of these systems. Several papers arc concerned with analyzing multiprocessor systems with non-uniform access probabilities without bus contention [13,14]. These studies are concerned with modeling local reference behavior.…”

Section: Introductionmentioning

confidence: 99%

An approximate analysis of multiprocessor systems

Towsley

1983

Proceedings of the 1983 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems - SIGMETRICS '83

View full text Add to dashboard Cite

This paper presents an approximate analysis of a multiprocessor system consisting of P processors, M memory modules, and B buses. The model assumes constant memory access times, arbitrary memory access patterns, and bus contention. The solution technique aggregates all memories into a composite queue and degrades the service rates of this queue so as to include the effect of bus contention. The throughput predictions from this model are very accurate, typically within 1% of predictions made with either simulation or exact analysis.

show abstract

Performance of Cross-Bar Multiprocessor Systems

Cited by 22 publications

References 9 publications

Simulation experiments for performance analysis of multiple-bus multiprocessor systems with nonexponential service times

Simulation experiments for performance analysis of multiple-bus multiprocessor systems with nonexponential service times

A comprehensive performance evaluation of crossbar networks

An approximate analysis of multiprocessor systems

Contact Info

Product

Resources

About