Analysis of polling systems
A polling system is one that contains a number of queues served in cyclic order. It is employed in computer-terminal communication systems and implemented in such standard data link protocols as BSC, SDLC, and HDLC, and its analysis is now finding a new application in local-area computer networks.This monograph analyzes polling systems to evaluate such basic performance measures as the average queue length and waiting time. Following a taxonomy of models with reference to previous work, it considers one-message buffer systems and infinite buffer systems with exhaustive, gated, and limited service disciplines. Examples to which the analysis of polling systems is applied are drawn from the field of computer communication networks. Contents: Introduction. One-Message Buffer Systems. Exhaustive Service, Discrete-Time Systems. Exhaustive Service, Continuous-Time Systems. Gated Service Systems. Limited Service Systems. Systems with Zero Reply Intervals. Sample Applications. Future Research Topics. Summary of Important Results.Hideaki Takagi is with IBM Japan Science Institute in Tokyo. Analysis of Polling Systems is included in the Computer Systems Series, Research Reports and Notes, edited by Herb Schwetman.
What people are saying - Write a review
We haven't found any reviews in the usual places.
ExhaustiveService DiscreteTime Systems
ExhaustiveService ContinuousTime Systems
Gated Service Systems
7 other sections not shown
arrival process arriving at station busy period Chapter conservation law consider continuous-time systems define denote derive discrete-time system distribution function exhaustive service system expression Expressnet gambler's ruin gated service system given identical stations IEEE instant for station intervisit joint GF l-Np Laplace transform limited service system Little's result mean and variance mean cycle mean message response mean message waiting mean number mean waiting message arrivals messages at station messages found messages served Networks nonidentical stations Note number of arrivals number of messages number of packets obtain one-message buffer system packets arriving packets at station parameters Poisson arrival polling cycle polling instant polling message polling system probability probability density function propagation delay Queueing Models Queueing Theory regenerative cycle Section server service completion slot solved station i+1 super cycle supermessage switch point tagged message th visit Ti(m token bus token ring zero reply intervals