Threshold-based congestion control scheme for ABR services in ATM networks

✍ Scribed by Jin-Long Wang

• Publisher: Elsevier Science
• Year: 2002
• Tongue: English
• Weight: 782 KB
• Volume: 25
• Category: Article
• ISSN: 0140-3664
• DOI: 10.1016/s0140-3664(01)00395-4

✦ Synopsis

In this paper, an effective ¯ow control scheme, modi®ed explicit rate indication for congestion avoidance (MERICA), is proposed to adaptively release congestion and enhance throughput for available bit rate (ABR) services in asynchronous transfer mode (ATM) networks. The proposed scheme mainly employs thresholds to determine the traf®c status. Whenever the traf®c status is changed, the switch will immediately inform the source of the new allowed data rate through the resource management (RM) cell in order to regulate the ABR traf®c. For ef®ciently deriving thresholds, a Markov chain model is developed to precisely describe the behavior of the traf®c pattern in ATM networks. Moreover, an algorithm is proposed to ef®ciently solve the thresholds from the Markov chain model under the time complexity of O(N). Finally, a simulation model is presented to investigate the performance. From the simulation results, the proposed scheme can obtain lower loss rate and higher utilization than other methods.