Design, Implementation and Performance of an Edge Computing Prototype Using Raspberry Pis
Tyler Gizinski, Xiang Cao
Abstract
Edge computing nowadays has been a novel computing paradigm, which intends to improve the performance of many applications. In edge computing, edge servers are usually deployed at locations that are closer to data sources, so that latency and response time can be reduced. Recently, Raspberry Pi has become a popular small low-cost computer that can be deployed for many Internet of Things (IoT) applications. In this paper, we design and implement an edge computing prototype using Raspberry Pi devices. We demonstrate the feasibility of edge computing with Raspberry Pis as edge servers. We take advantage of the local and parallel processing capabilities of Raspberry Pis to show the scenarios where edge computing can outperform the traditional cloud computing. Hands-on experiments verify that in some cases, edge computing has better performance in terms of total finishing time of applications.