Task Scheduling Algorithm Selection in Cloud Computing Through Machine Learning Techniques

The cloud’s popularity expanded dramatically because of the cheaper costs and greater flexibility cloud service providers offer. The cloud service provider has to schedule requests that come in dynamically. Task scheduling is necessary in a cloud environment to guarantee effective resource use. Task scheduling is therefore essential to the effectiveness and efficiency of cloud computing techniques. Although there are many ways to make work scheduling in the cloud better, the problem is yet unsolved. This study introduces a proposed model for optimizing cloud computing resource usage through the use of machine learning algorithms. The model discussed in this work considers key scheduling factors such as cost, throughput, and makespan. To arrive at an algorithm most relevant for a specific scheduling criterion, the model incorporates a machine learning model with four heuristic algorithms: Min-min, Max-min, Simulated Annealing, and Sufferage. In addition, multi-linear regression is utilized in the model to discern an algorithm most closely mapping a predefined set of factors concerning a task. The outcomes of the experiment show notable gains in scheduling effectiveness, resource usage, and system performance.