Comparative evaluation of resource allocation strategies using weighted round robin scheduler in ATM switches

We study resource allocation strategies in ATM switches, which provide qual ity of service (QoS) guarantees to individual connections. In order to mini mize the cell loss rate over a wide range of traffic characteristics, an ef ficient allocation strategy is necessary. in this paper we introduce a reso urce allocation strategy, named TP+WRR (Threshold Pushout + Weighted Round Robin) which can fully utilize the buffer space and the bandwidth. We compa re the performance of TP+WRR with two typical resource allocation strategie s. An exact queueing analysis based on a Markov model is carried out under bursty traffic sources to evaluate their performance. Our results reveal th at TP+WRR considerably improves the cell loss probability over the other st rategies considered in this paper, especially when many connections are sha ring a link.