A Continuous Overlay Path Probing Algorithm For Overlay Networks

Bandwidth is a key factor in network technologies and it has been of major importance throughout the history of packet networks. In fact, bandwidth estimation is very beneficial to optimize the performance of end-to-end transport in several overlay applications such as Content Distribution Networks...

Full description

Bibliographic Details
Main Author: Feily, Maryam
Format: Thesis
Language:English
Published: 2013
Subjects:
Online Access:http://eprints.usm.my/45118/
http://eprints.usm.my/45118/1/Maryam%20Feily24.pdf
_version_ 1848880237714079744
author Feily, Maryam
author_facet Feily, Maryam
author_sort Feily, Maryam
building USM Institutional Repository
collection Online Access
description Bandwidth is a key factor in network technologies and it has been of major importance throughout the history of packet networks. In fact, bandwidth estimation is very beneficial to optimize the performance of end-to-end transport in several overlay applications such as Content Distribution Networks (CDNs), Peer-to-Peer (P2P) file sharing, and dynamic overlay routing. The end-to-end available bandwidth determines the extra bandwidth that can be provided to overlay traffic. Knowledge about the available bandwidth of an overlay path enables dynamic rate adoption and better bandwidth utilization by content distribution schemes in overlay networks. However, the important issue is how to measure the available bandwidth on an end-to-end overlay path without prior knowledge about the physical network. Over the last two decades, researchers have been trying to create algorithms to measure end-to-end available bandwidth and other bandwidth-related metrics accurately, quickly, and without affecting the traffic of the path. Active measurement techniques performed by overlay nodes can provide bandwidth estimations of an end-to-end overlay path. This thesis describes a new algorithm called “COPPA,” which is an in-band path probing algorithm for measuring the end-to-end available bandwidth of an overlay path accurately and continuously. The aim is to provide up-to-date bandwidth information for enhanced content distribution processes in overlay networks. The primary idea is to perform active measurements using the applications’ packets instead of using extra probe packets. Such an in-band probing algorithm reduces measurement overhead on the selected overlay path. Several experiments were carried out using the OMNeT++ simulation framework. The designed algorithm was evaluated using experimental data. The obtained results show that the continuous in-band overlay path probing algorithm (COPPA) provides up-to-date bandwidth information with reduced overhead and minimal impact on the traffic of the path.
first_indexed 2025-11-15T17:59:56Z
format Thesis
id usm-45118
institution Universiti Sains Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T17:59:56Z
publishDate 2013
recordtype eprints
repository_type Digital Repository
spelling usm-451182019-07-31T08:37:10Z http://eprints.usm.my/45118/ A Continuous Overlay Path Probing Algorithm For Overlay Networks Feily, Maryam QA75.5-76.95 Electronic computers. Computer science Bandwidth is a key factor in network technologies and it has been of major importance throughout the history of packet networks. In fact, bandwidth estimation is very beneficial to optimize the performance of end-to-end transport in several overlay applications such as Content Distribution Networks (CDNs), Peer-to-Peer (P2P) file sharing, and dynamic overlay routing. The end-to-end available bandwidth determines the extra bandwidth that can be provided to overlay traffic. Knowledge about the available bandwidth of an overlay path enables dynamic rate adoption and better bandwidth utilization by content distribution schemes in overlay networks. However, the important issue is how to measure the available bandwidth on an end-to-end overlay path without prior knowledge about the physical network. Over the last two decades, researchers have been trying to create algorithms to measure end-to-end available bandwidth and other bandwidth-related metrics accurately, quickly, and without affecting the traffic of the path. Active measurement techniques performed by overlay nodes can provide bandwidth estimations of an end-to-end overlay path. This thesis describes a new algorithm called “COPPA,” which is an in-band path probing algorithm for measuring the end-to-end available bandwidth of an overlay path accurately and continuously. The aim is to provide up-to-date bandwidth information for enhanced content distribution processes in overlay networks. The primary idea is to perform active measurements using the applications’ packets instead of using extra probe packets. Such an in-band probing algorithm reduces measurement overhead on the selected overlay path. Several experiments were carried out using the OMNeT++ simulation framework. The designed algorithm was evaluated using experimental data. The obtained results show that the continuous in-band overlay path probing algorithm (COPPA) provides up-to-date bandwidth information with reduced overhead and minimal impact on the traffic of the path. 2013-07 Thesis NonPeerReviewed application/pdf en http://eprints.usm.my/45118/1/Maryam%20Feily24.pdf Feily, Maryam (2013) A Continuous Overlay Path Probing Algorithm For Overlay Networks. PhD thesis, Universiti Sains Malaysia.
spellingShingle QA75.5-76.95 Electronic computers. Computer science
Feily, Maryam
A Continuous Overlay Path Probing Algorithm For Overlay Networks
title A Continuous Overlay Path Probing Algorithm For Overlay Networks
title_full A Continuous Overlay Path Probing Algorithm For Overlay Networks
title_fullStr A Continuous Overlay Path Probing Algorithm For Overlay Networks
title_full_unstemmed A Continuous Overlay Path Probing Algorithm For Overlay Networks
title_short A Continuous Overlay Path Probing Algorithm For Overlay Networks
title_sort continuous overlay path probing algorithm for overlay networks
topic QA75.5-76.95 Electronic computers. Computer science
url http://eprints.usm.my/45118/
http://eprints.usm.my/45118/1/Maryam%20Feily24.pdf