Opportunistic Deployment of Distributed Edge Clouds for Latency-critical Applications

Abstract

The growing number of latency-critical applications are posing novel challenges for network operators, cloud/hosting companies, and application providers. Edge Computing is the strongest candidate for providing low-latency responses, but it is not yet clear what edge infrastructures will be like. This paper introduces a new platform for enabling an edge infrastructure according to a disaggregated distributed cloud architecture and an opportunistic model based on bare-metal providers. Results from a multi-server online gaming application deployed in a real geo-distributed edge infrastructure show the feasibility, performance and cost efficiency of the solution.