Server selection and admission control for IP-based video on demand using available bandwidth estimation

Brian Meskill, Alan Davy, Brendan Jennings

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

Service providers offering IP-based video on demand services often replicate video content in multiple content servers with different network points of attachment. When a request for a content item arrives from an end-user, a decision must be made as to whether the request should be admitted and, if so, which server should be used. To ensure adequate quality-of-service this admission control/server selection decision should be cognisant of the current link utilisation on the paths between the end-user's point of attachment and those of the servers. We have studied the conditions and parameters under which end-to-end available bandwidth estimation tools (ABETs) pathChirp and Assolo can be used in this decision process. We specify an admission control/server selection algorithm that uses available bandwidth estimations. Simulation study results show pathChirp (but not Assolo), when appropriately parameterised, can generate available bandwidth estimates that can be used by our admission control/server selection algorithm to react appropriately to changes in background loads on network paths.

Original languageEnglish
Title of host publicationProceedings of the 36th Annual IEEE Conference on Local Computer Networks, LCN 2011
Pages255-258
Number of pages4
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event36th Annual IEEE Conference on Local Computer Networks, LCN 2011 - Bonn, Germany
Duration: 4 Oct 20117 Oct 2011

Publication series

NameProceedings - Conference on Local Computer Networks, LCN

Conference

Conference36th Annual IEEE Conference on Local Computer Networks, LCN 2011
Country/TerritoryGermany
CityBonn
Period4/10/117/10/11

Fingerprint

Dive into the research topics of 'Server selection and admission control for IP-based video on demand using available bandwidth estimation'. Together they form a unique fingerprint.

Cite this