For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Achieving Max-Min Fairness by Decentralization for the ABR Traffic Control in ATM Networks
Seung Hyong RHEE Takis KONSTANTOPOULOS
IEICE TRANSACTIONS on Communications
Publication Date: 2001/08/01
Print ISSN: 0916-8516
Type of Manuscript: PAPER
ATM network, available bit rate, decentralized control, max-min fairness, equilibrium,
Full Text: PDF>>
The available bit rate (ABR) is an ATM service category that provides an economical support of connections having vague requirements. An ABR session may specify its peak cell rate (PCR) and minimum cell rate (MCR), and available bandwidth is allocated to competing sessions based on the max-min policy. In this paper, we investigate the ABR traffic control from a different point of view: Based on the decentralized bandwidth allocation model studied in , we prove that the max-min rate vector is the equilibrium of a certain system of noncooperative optimizations. This interpretation suggests a new framework for ABR traffic control that allows the max-min optimality to be achieved and maintained by end-systems, and not by network switches. Moreover, in the discussion, we consider the constrained version of max-min fairness and develop an efficient algorithm with theoretical justification to determine the optimal rate vector.