Queueing Network Modeling of Computer Communication Networks
ACM Computing Surveys (CSUR)
Linearizer: a heuristic algorithm for queueing network models of computing systems
Communications of the ACM
Computer Performance Modeling Handbook
Computer Performance Modeling Handbook
Multiple class memory constrained queueing networks
SIGMETRICS '82 Proceedings of the 1982 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Fast approximate solution of multiprogramming models
SIGMETRICS '82 Proceedings of the 1982 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Analysis of Queueing Network Models with population size constraints and delayed blocked customers
SIGMETRICS '84 Proceedings of the 1984 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Solving Semi-Open Queuing Networks
Operations Research
| Hi-index | 0.00 |
A method for analyzing the performance of computer communication networks using queueing network models (QNM's) is presented in this paper. The user/computer sites are interconnected by a wide-area network with an end-to-end window flow-control mechanism. The QNM for the system including the communication subnet and the host computers is intractable because of blocking due to window flow control and memory constraints at computer systems, even under otherwise analytically favorable assumptions. The approximate solution method, which is based on iteration and decomposition, combines several recently developed computationally inexpensive algorithms to compute user response times, taking into account the delay in the communication network and the processing time at computer systems. We also introduce a fast non-iterative procedure to analyze QNM's with external arrivals and with population size constraints, such that blocked customers are queued. The effect of the choice of window sizes and minimum link/bifurcated routing on system performance is investigated.