Edge Computing vs Centralized Cloud: Impact of Communication Latency on the Energy Consumption of LTE Terminal Nodes

TL;DR

This paper investigates how communication latency differences between edge and cloud computing affect LTE terminal energy consumption, revealing that edge servers significantly reduce energy use especially under certain data transfer conditions.

Contribution

It provides an empirical analysis of energy consumption in LTE nodes communicating with edge versus cloud servers, highlighting the impact of latency and data transfer direction.

Findings

Edge servers reduce energy consumption compared to cloud servers.

Energy savings depend on data transfer direction, server load, and time of day.

Reduced latency at edge servers offers significant energy benefits.

Abstract

Edge computing brings several advantages, such as reduced latency, increased bandwidth, and improved locality of traffic. One aspect that is not sufficiently understood is to what extent the different communication latency experienced in the edge-cloud continuum impacts on the energy consumption of clients. We studied the energy consumption of a request-response communication scheme when an LTE node communicates with edge-based or cloud-based servers. Results show that the reduced latency of edge servers bring significant benefits in terms of energy consumption. Experiments also show how the energy savings brought by edge computing are influenced by the prevalent direction of data transfer (upload vs download), load of the server, and daytime/nighttime operation.