Recursive relations are derived for the exact computation of the steady-state probability distribution of some queuing models with passive resources that can be used to analyze the performance of multiple-bus multiprocessor system architectures. The most general case that can be shown to admit a product-form solution is described, and a recursive solution is obtained taking into account, considering different processor access rates, different memory selection probabilities, and a first-come-first-served bus scheduling policy. Several simpler cases allowing easier model solutions are also considered. Numerical evaluations for large computing systems with nonuniform memory references show the usefulness of the results

Product-Form Solution Techniques for the Performance Analysis of Multiple-Bus Multiprocessor Systems with Nonuniform Memory References

BALBO, Gianfranco
1988-01-01

Abstract

Recursive relations are derived for the exact computation of the steady-state probability distribution of some queuing models with passive resources that can be used to analyze the performance of multiple-bus multiprocessor system architectures. The most general case that can be shown to admit a product-form solution is described, and a recursive solution is obtained taking into account, considering different processor access rates, different memory selection probabilities, and a first-come-first-served bus scheduling policy. Several simpler cases allowing easier model solutions are also considered. Numerical evaluations for large computing systems with nonuniform memory references show the usefulness of the results
1988
37 (5) May 1988
532
540
exact computation; first-come-first-served bus scheduling policy; local balance property; memory selection probabilities; multiple-bus multiprocessor systems; multiprocessor interconnection networks; nonuniform memory references; performance evaluation; processor access rates; product-form solution; queueing theory; queuing models with passive resources; recursive solution; scheduling.; steady-state probability distribution
G. CHIOLA; M. AJMONE MARSAN; G. BALBO
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/26090
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact