A model of bandwidth utilization in wide-band computer communication networks

Abstract

A wide-band integrated service digital network provides access to many different users in a shared fashion. Control of the available bandwidth allocation is necessary for efficiency and flexibility. The system is modeled as a multiple-server multiple-queue system, where access requests (calls) require service from a random number of servers. Each call is entered in a queue so that the queue index indicates the bandwidth needed by the call (in basic bandwidth units). Control of service allocation is accomplished by assigning priorities to queues. A variety of priority types are examined, where relatively higher priority is assigned to calls in the smaller indexed or the larger indexed queues. When more than one call can be served due to the availability of bandwidth, the one with the highest priority is chosen first. The model takes into account the queue selection probabilities and expected service times which are assumed to depend on the queue index. Also, the effect of queue capacity is examined. The procedures that regulate the bandwidth allocation are described, and a Markov-chain based solution to the model is developed to produce the system state probabilities, and utilization and performance measures.