THE STICKY BUFFER - FLOW-CONTROL FOR THE ATM

Citation
G. Hjalmtysson et Ag. Konheim, THE STICKY BUFFER - FLOW-CONTROL FOR THE ATM, Performance evaluation, 26(4), 1996, pp. 289-320
Citations number
16
Categorie Soggetti
Computer Sciences","Computer Science Hardware & Architecture","Computer Science Theory & Methods
Journal title
ISSN journal
01665316
Volume
26
Issue
4
Year of publication
1996
Pages
289 - 320
Database
ISI
SICI code
0166-5316(1996)26:4<289:TSB-FF>2.0.ZU;2-K
Abstract
Data from different applications - voice, video, file transfer, intera ctive - will be multiplexed in the future over broadband integrated se rvices digital networks (BISDN). Data are segmented into 48-byte block s prefixed by a 5-byte header and transported over the network using t he asynchronous transfer mode (ATM). An ATM connection-request is a co ntract between the user and the network; the user specifies a rate req uirement, delay constraints, a bound on the cell-loss probability and other quality of service parameters, If the network can meet these req uirements, a connection is made. Bursty traffic producing peak traffic rates in excess of the projected average rate could result in congest ion and lead to performance degradation. As a result, the network migh t no longer be able to deliver the negotiated quality of service to ex isting connections. To lessen the chance of congestion, input rate con trol must be implemented. This paper contains a performance analysis o f the sticky buffer (SE), a moving window input rate control scheme in tended to limit the rate at which input traffic may enter a network. P olicing is achieved by buffering and hence delaying the entry of ATM-c ells into the network. The scheme is determined by parameters (R, T) w hich specify that no more than R ATM-cells are permitted to enter the network in every window of size T cells. We present an exact analysis, deriving the probability generating function of the queue length dist ribution. A comparison with the leaky bucket is given. Our numerical e xamples show the required buffer size is comparable to that required b y the leaky bucket.