"An Experimental study on the impact of execution location in edge-clou" by Dimitrios Melissourgos, Sishun Wang et al.
 

Computer Science

An Experimental study on the impact of execution location in edge-cloud computing

Dimitrios Melissourgos, Herbert Wertheim College of Engineering
Sishun Wang, Herbert Wertheim College of Engineering
Shigang Chen, Herbert Wertheim College of Engineering
Youlin Zhang, Herbert Wertheim College of Engineering
Olufemi Odegbile, Herbert Wertheim College of Engineering
Yuanda Wang, Herbert Wertheim College of Engineering

Abstract

On the one hand, edge computing has the advantage of distributing the load to the edges of a computer network. Local computation at the edge is bandwidth-efficient and anonymous. On the other hand, cloud computing is the choice when it comes to computationally demanding tasks and big data. In this paper, we argue for edge-cloud computing (which blends the two together) with an experimental study on the impact of execution location on application performance. We answer the question of how to determine whether it should compute a task at the edge or on the cloud and what the criteria are. We analyze the factors of response time, memory space, data availability and privacy policy. We experimentally evaluate the impact of these factors on execution location based on a network visualizer software.