A Geo-Aware Server Assignment Problem for Mobile Edge Computing

TL;DR

This paper addresses the challenge of geographically placing edge servers in mobile edge computing to optimize user experience and application performance, especially for pairwise communication between mobile devices.

Contribution

It introduces a new formulation for the geo-aware server assignment problem and evaluates heuristic solutions using real-world and synthetic data.

Findings

Heuristic solutions improve server placement efficiency.

The proposed formulation effectively enhances application performance.

Evaluation demonstrates benefits over baseline approaches.

Abstract

As mobile devices have become the preferred tool for communication, work, and entertainment, traffic at the edge of the network is growing more rapidly than ever. To improve user experience, commodity servers are deployed in the edge to form a decentralized network of mini datacenters each serving a localized region. A challenge is how to place these servers geographically to maximize the offloading benefit and be close to the users they respectively serve. We introduce a formulation for this problem to serve applications that involve pairwise communication between mobile devices at different geolocations. We explore several heuristic solutions and compare them in an evaluation using both real-world and synthetic datasets.