AVERAGE WAITING TIME PROFILES OF UNIFORM DISTRIBUTED QUEUE DUAL BUS SYSTEM MODEL

Citation
Nsv. Rao et al., AVERAGE WAITING TIME PROFILES OF UNIFORM DISTRIBUTED QUEUE DUAL BUS SYSTEM MODEL, IEEE transactions on parallel and distributed systems, 6(10), 1995, pp. 1068-1084
Citations number
41
Categorie Soggetti
System Science","Engineering, Eletrical & Electronic","Computer Science Theory & Methods
ISSN journal
10459219
Volume
6
Issue
10
Year of publication
1995
Pages
1068 - 1084
Database
ISI
SICI code
1045-9219(1995)6:10<1068:AWTPOU>2.0.ZU;2-5
Abstract
The Distributed Queue Dual Bus (DQDB) system consists of a linear arra ngement of N nodes that communicate with each other using two contra-f lowing buses, The nodes use an extremely simple protocol to send messa ges on these buses. This simple, but elegant, system has been found to be very challenging to analyze. We consider a simple and uniform abst raction of this model to highlight the fairness issues in terms of ave rage waiting time. We introduce a new approximation method to analyze the performance of DQDB system in terms of the average waiting time of a node expressed as a function of its position. Our approach abstract s the intimate relationship between the load of the system and its fai rness characteristics, and explains all basic behavior profiles of DQD B observed in previous simulation. For the uniform DQDB with equal dis tance between adjacent nodes, we show that the system operates under t hree basic behavior profiles and a finite number of their combinations that depend on the load of the network, Consequently, the system is n ot fair at any load in terms of the average waiting times. We also sho w that the main theme of the analysis carries over to the general (non uniform) DQDB, By suitably choosing the inter-node distances, the DQDB can be made fair around some loads, but such system will become unfai r as the load changes, In the vicinity of a critical toad, the uniform network runs into a state of instability, where its behavior fluctuat es from one extreme to the other with small load variations, Our analy sis is supported by simulation results.