King's College London

Research portal

Consume Local: Towards Carbon Free Content Delivery

Research output: Chapter in Book/Report/Conference proceedingConference paper

Aravindh Raman, Dmytro Karamshuk, Nishanth Sastry, Andrew Secker, Jigna Chandaria

Original languageEnglish
Title of host publicationInternational Conference on Distributed Computing Systems
PublisherIEEE
StatePublished - Jul 2018

Documents

  • Consume Local: Towards Carbon_RAMAN_2018_GREEN AAM

    ICDCS2018_consume_local_1_.pdf, 417 KB, application/pdf

    6/06/2018

    Accepted author manuscript

    Other

    © 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

King's Authors

Abstract

P2P sharing amongst consumers has been proposed as a way to decrease load on Content Delivery Networks. This paper develops an analytical model that shows an additional benefit of sharing content locally: Selecting close by peers to share content from leads to shorter paths compared to traditional CDNs, decreasing the overall carbon footprint of the system. Using data from a month-long trace of over 3 million monthly users in London accessing TV shows online, we show that local sharing can result in a decrease of 24-48% in the system-wide carbon footprint of online video streaming, despite various obstacle factors that can restrict swarm sizes. We confirm the robustness of the savings by using realistic energy parameters drawn from two widely used settings. We also show that if the energy savings of the CDN servers are transferred as carbon credits to the end users, over 70% of users can become carbon positive, i.e., are able to support their content consumption without incurring any carbon footprint, and are able to offset their other carbon consumption. We suggest carbon credit transfers from CDNs to end users as a novel way to incentivise participation in peer-assisted content delivery.

Download statistics

No data available

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454