A Predictive Mechanism for ABR Traffic Management in ATM Networks

Abstract

We propose a predictive mechanism for managing the ABR traffic ATM networks. For ABR traffic management, the establishment of a feedback loop between the source/destination end systems , and the network lets the data sources know the state of the networks, and utilize the available network bandwidth more efficiently. However, because of the propagation delay of sending back the resource management cells to the data sources, the networks is either underutilized (when the network load is light), or congested ( when the network load is heavy) by the excessive cells that enter the network during the propagation period. The proposed predictive mechanism based on Trellis diagram predicts the contents of the resource management cell and makes rate adjustment beforehand. We also show how the predictive mechanism can be used as an “add-on” to the current congestion control algorithms such as EFCI, PRCA, and EPRCA. A performance evaluation demonstrates that the predictive approach reduces excessive cells and increases network utilization.