Extensions and computational aspects of an iterative method

Authors:
Raymond A. Marie;Patricia M. Snyder;William J. Stewart
Affiliations:
North Carolina State University, Raleigh, N. C.;North Carolina State University, Raleigh, N. C.;North Carolina State University, Raleigh, N. C.
Venue:
SIGMETRICS '82 Proceedings of the 1982 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Year:
1982

Citing 5
Cited 2

Application of the Diffusion Approximation to Queueing Networks I: Equilibrium Queue Distributions

Journal of the ACM (JACM)
Open, Closed, and Mixed Networks of Queues with Different Classes of Customers

Journal of the ACM (JACM)
On Approximate Computer System Models

Journal of the ACM (JACM)
Calculating equilibrium probabilities for &lgr;(n)/Ck/1/N queues

PERFORMANCE '80 Proceedings of the 1980 international symposium on Computer performance modelling, measurement and evaluation
HAM: the heuristic aggregation method for solving general closed queueing network models of computer systems

SIGMETRICS '82 Proceedings of the 1982 ACM SIGMETRICS conference on Measurement and modeling of computer systems

Approximate Analysis of Load Dependent General Queueing Networks

IEEE Transactions on Software Engineering
Nonexponential networks of queues: a maximum entropy analysis

SIGMETRICS '85 Proceedings of the 1985 ACM SIGMETRICS conference on Measurement and modeling of computer systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The so-called iterative methods are among a class of methods that have recently been applied to obtain approximate solutions to general queueing networks. In this paper it is shown that if the network contains feedback loops, then it is more advantageous to incorporate these loops into the analysis of the station itself rather than into the analysis of the complement of the station. We show how this analysis may be performed for a simple two-phase Coxian server. Additionally, it is shown that the number of iterations required to achieve a specified degree of accuracy may be considerably reduced by using a continuous updating procedure in which the computed throughputs are incorporated as soon as they are available, rather than at the end of an iteration. An efficient computational scheme is presented to accompany this continuous updating. Finally a number of examples are provided to illustrate these features.