MUTE: MUlti-Tier Edge Networks

Alessio Silvestro, Nitinder Mohan, Jussi Kangasharju, Fabian Schneider, Xiaoming Fu

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

Abstract

Several Internet-of-Things (IoT) based use-cases require an increased amount of computation resources with extremely low latency. Cloud-based services may fail to provide such requirement given the high latency required to access their facilities. Therefore, an increasing number of resources are being deployed at the network "edge" as Edge Clouds.

However, as the modeling of edge network is still in its early stages, the existing solutions for service placement and resource utilization are found to be quite inefficient. In this paper, we model a multi-tier edge network and propose a service placement algorithm, Mute.
In our evaluation, performed on real network topologies, we show that Mute achieves 66% reduction in network cost and 50% reduction in service placement when compared to state-of-the-art solutions.
Original languageEnglish
Title of host publicationCrossCloud'18: Proceedings of the 5th Workshop on CrossCloud Infrastructures & Platforms
Number of pages6
Place of PublicationNew York
PublisherACM
Publication date23 Apr 2018
ISBN (Electronic)978-1-4503-5653-4
DOIs
Publication statusPublished - 23 Apr 2018
MoE publication typeA4 Article in conference proceedings
EventWorkshop on CrossCloud Infrastructure and Platforms - Porto, Portugal
Duration: 23 Apr 201826 Apr 2018
Conference number: 5
http://www.comp.lancs.ac.uk/~elkhatib/crosscloud/

Fields of Science

  • 113 Computer and information sciences

Cite this