Stability of rate-based congestion control algorithm for ABR service class in ATM networks
Hiroyuki Ohsaki, Masayuki Murata, Hideo Miyahara
Abstract
Hiroyuki Ohsaki, Masayuki Murata, Hideo Miyahara
Abstract
A rate-based congestion control algorithm is a feedback-based flow control mechanism for ABR (available bit rate) service class, which is suitable for data transfer applications. The rate-based congestion control algorithm dynamically regulates a cell transmission rate of a source end system according to feedback information from the network. An appropriate choice of control parameters can lead to effective use of network resources. However, once it fails, the rate-based congestion control unexpectedly exhibits unstable operation under some parameter settings, which leads to dramatic degradation of the performance. We investigate causes of unstable operation and performance degradation of the rate-based congestion control. We then propose an improvement in the existing rate-based congestion control algorithm. Its effectiveness is demonstrated by providing several simulation experiments.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A rate-based congestion control algorithm is a feedback-based flow control mechanism for ABR (available bit rate) service class, which is suitable for data transfer applications. The rate-based congestion control algorithm dynamically regulates a cell transmission rate of a source end system according to feedback information from the network. An appropriate choice of control parameters can lead to effective use of network resources. However, once it fails, the rate-based congestion control unexpectedly exhibits unstable operation under some parameter settings, which leads to dramatic degradation of the performance. We investigate causes of unstable operation and performance degradation of the rate-based congestion control. We then propose an improvement in the existing rate-based congestion control algorithm. Its effectiveness is demonstrated by providing several simulation experiments.
Key concepts: Network congestion, Flow control (data), Computer science, Asynchronous Transfer Mode, Computer network, Transmission (telecommunications), Degradation (telecommunications), Bit rate