Edge Computing Vs. Cloud Computing: Evaluating Performance, Scalability, and Security in Modern Applications

Abstract

Emerging computing paradigms such as edge computing and cloudlet computing offer distributed computing resources in proximity to the sources and end users. By moving the processing closer to the data source and consumer, these novel paradigms improve the performance by decreasing latencies, and thus by enabling a faster and real-time provision of services. In addition, they can be used to offload massive amounts of raw data for processing on more powerful computing resources instead of sending all the data through the network towards the clouds, thus saving network bandwidth and mitigating congestion. These computing paradigms also enable more flexible service provision, as the resources can be easily scaled based on the demand in a productive and cost-effective way compared to the traditional cloud computing infrastructures. The rapid evolution of modern technologies has led to the emergence of a myriad of applications in different domains such as Industry 4.0, the Internet of Things, smart cities, health-care, and transportation. The typical processing approach in these applications involves data travel from sources to machines or servers for processing and then back to the end users. Since the data travel distance might be long depending on network and topological structures, traditional cloud computing is found to be insufficient to fulfil the performance, scalability, and security requirements of many modern applications. Therefore, there is an urgent need for novel computing paradigms which close the processing gap between data sources and consumers.