Queues and packet multiplexing networks

This thesis has to do with certain fundamental queues that are well established as models for delay in simple packet-switching concentrators and networks. We first revisit the single server queue with Poisson arrivals and general independent service times. We then work out a complete delay analysis for a traffic concentrating tandem network of queues with deterministic service and batch Poisson sources connected to every node; this is the most comprehensive analysis available for a network which is not of Jackson type. We also show how to (partially) extend the analysis to a concentrating tree network, and to an arrival process somewhat more general that batch Poisson. / The two parts of the thesis have a close methodological relationship. Our contribution in both cases is to rederive certain known results, and to produce a variety of new ones, using techniques that are essentially qualitative. Our particular view of the stochastic processes in question is guided by a very special queue discipline, namely Last Come First Served preemptive resume; by identifying certain structural features of the sample paths, one can read, almost without calculation, a host of statistics of common interest. The LCFS preemptive resume discipline also enables us: (i) to strengthen the connection between the single server queue with general independent service times and interarrival times, and the fluctuation theory of random walks; (ii) to strengthen the connection between the queue with Poisson arrivals and branching processes.

Identiferoai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.71999
Date January 1985
CreatorsShalmon, Michael S.
PublisherMcGill University
Source SetsLibrary and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada
LanguageEnglish
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Formatapplication/pdf
CoverageDoctor of Philosophy (Department of Electrical Engineering.)
RightsAll items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated.
Relationalephsysno: 000219546, proquestno: AAINL20863, Theses scanned by UMI/ProQuest.

Page generated in 0.002 seconds