Resource usage prediction algorithms for optimal selection of multimedia content delivery methods

Yiannos Kryftis, Constandinos X. Mavromoustakis, George Mastorakis, Evangelos Pallis, Jordi Mongay Batalla, Joel J.P.C. Rodrigues, Ciprian Dobre, Georgios Kormentzas

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

Abstract

This paper proposes two algorithms adopted in a prototype network architecture, for optimal selection of multimedia content delivery methods, as well as balanced delivery load, by exploiting a novel resource prediction engine. The proposed architecture exploits both algorithms for the prediction of future multimedia services demands, by providing the ability to keep optimal the distribution of the streaming data, among Content Delivery Networks, cloud-based providers and Home Media Gateways. In addition, the prediction of the upcoming fluctuations of the network, provides the ability to the proposed network architecture, achieving optimized Quality of Service (QoS) and Quality of Experience (QoE) for the end users. Both algorithms were evaluated to establish their efficiency, towards effectively predicting future network traffic demands. The experimental results validated their performance and indicated fields for further research and experimentation.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Communications, ICC 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages5903-5909
Number of pages7
Volume2015-September
ISBN (Electronic)9781467364324
DOIs
Publication statusPublished - 9 Sept 2015
EventIEEE International Conference on Communications, ICC 2015 - London, United Kingdom
Duration: 8 Jun 201512 Jun 2015

Other

OtherIEEE International Conference on Communications, ICC 2015
Country/TerritoryUnited Kingdom
CityLondon
Period8/06/1512/06/15

Keywords

  • Content Delivery Networks
  • Multimedia Services Systems
  • Resource Prediction Engine
  • Resource Usage Prediction Algorithms

Fingerprint

Dive into the research topics of 'Resource usage prediction algorithms for optimal selection of multimedia content delivery methods'. Together they form a unique fingerprint.

Cite this