Active loss pair measurement

Pao Yue-kong Library Electronic Theses Database

Active loss pair measurement

 

Author: Fok, Wai-ting
Title: Active loss pair measurement
Degree: M.Sc.
Year: 2008
Subject: Hong Kong Polytechnic University -- Dissertations.
Routers (Computer networks)
Computer networks -- Evaluation.
Department: Dept. of Computing
Pages: 77 leaves : ill. ; 30 cm.
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b2240171
URI: http://theses.lib.polyu.edu.hk/handle/200/3557
Abstract: Measurement based on packet pairs has been used to discover end-to-end network characteristics. One of such measurement is loss-pair analysis [11], which has been shown to be able to passively measure the queue size in the bottleneck router along a path in ns-2 simulator. However, the methodology only models routers for which the queue is managed in the unit of byte. The application of the method is thus limited. In this paper, we have proposed an active measurement methodology that can apply to all routers. It make use of our newly proposed model that relates router queue size in unit of packet, bandwidth, packet size and queuing delay. We have performed experiments in testbed to validate our model. We have illustrated that our methodology can estimate queue size of bottleneck router which manages buffer in unit of packet. Finally, we have applied our methodology in a more complex baseline traffic. We found that if small-size packets dominates the base-line traffic, our methodology cannot give accurate results. We suggested some methods to increase the accuracy. The major contribution of this research are: 1. We have proposed an active loss-pair measurement methodology to support routers of different implementations of buffer management, such that the application can be extended to all routers in real world environment; 2. We have performed testbed experiment to verify the relationship between queue size of bottleneck router and queuing delay observed by a loss-pair; 3. We have demonstrated the use of linear regression and different statistical measures of central tendency to correct the error from data collected under different network traffic characteristics.

Files in this item

Files Size Format
b22401714.pdf 2.189Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information