Please use this identifier to cite or link to this item:
Title: Providing service differentiation in OBS networks through probabilistic preemption
Keywords: Quality of Service, Probabilistic Preemption, Blocking Probability, Blocking Probability Ratio
Issue Date: 20-Jul-2004
Citation: YANG LIHONG (2004-07-20). Providing service differentiation in OBS networks through probabilistic preemption. ScholarBank@NUS Repository.
Abstract: Optical Burst Switching (OBS) provides all-optical, high-speed data rate, and format transparent switching. In this thesis, a novel scheme, called Probabilistic Preemptive scheme, is proposed and investigated for achieving service differentiation in such networks.The main idea behind the Probabilistic Preemptive scheme is that a probabilistic parameter is added to the existing preemptive scheme. With this scheme, a high priority burst may preempt a low priority burst based on its probabilistic parameter to release the transmission period to the high priority burst. By changing the probabilistic parameter, the Probabilistic Preemptive scheme can achieve flexible blocking differentiation. Various aspects of the proposed Probabilistic Preemptive scheme are studied in this thesis. First, for the two-class system, the impact of preemptive probability on blocking probabilities of the two classes and the blocking ratio of two classes are investigated. Second, for the multiple class system, the three classes system is studied as an example. The impact of preemptive probability on blocking probabilities of the three classes is analysed. Third, for the multi-node system, the signalling issue is discussed. A release header is used to remove the switching information of the preempted burst in the downstream nodes of the network. The performance of a two-class system with or without releaser header is investigated.
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Thesis_(Yang Lihong).pdf551.11 kBAdobe PDF



Page view(s)

checked on Dec 16, 2018


checked on Dec 16, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.