Publication

Abstract : Today’s era is of intelligent devices connected to serve various real-time applications. These numerous IoT devices collect enormous amounts of data, which are then transferred to the cloud for further processing and analysis. Most of these tasks are delay-sensitive and require processing within their deadline. These tasks require high-speed computing and storage resources for their processing. Although the cloud layer provides enormous processing and storage capability, we can not ignore the amount of delay in response generation. Thus, other two lower layers, i.e., the Edge layer and the fog layer, should also be considered to develop an efficient algorithm for task scheduling and load balancing. These three layers vary in terms of resource availability and communication latency. Currently, the increased focus of many new researchers on improving the performance metrics in the cloud architecture, such as energy efficiency, resource utilization, and latency, leads to the requirement of a detailed review of previous work done in this area. Although many review articles are on the web, most focus only on a single layer. Thus, The current study considers all three layers, edge, fog, and cloud, to present a detailed systematic review of algorithms based on load balancing and scheduling at task and VM levels. This study aims to compare and analyze the available papers based on the focused layers, the methodology used, performance matrices, advantages, and future scopes.