Bivariate Correlative Modest Adaptive Boost Resource Aware Task Scheduling in Cloud Environment

Abstract

 Cloud computing is a rapidly evolving paradigm that provides accessible and virtualized resources through Internet technology. In this model, Cloud Service Providers (CSPs) offer online access to computing resources for users to execute their application tasks. Task scheduling in cloud computing involves the allocation of computational tasks to available resources within the cloud environment. The primary objectives of task scheduling are to optimize resource utilization, minimize task completion time, and enhance overall system performance. Task scheduling plays a vital role in cloud resource management as it directly impacts the efficiency of cloud data centers. With the increasing number of cloud computing users, task scheduling has become more challenging, requiring the use of suitable scheduling algorithms. To improve task scheduling efficiency, a novel method called Bivariate Correlative Modest Adaptive Boost Resource Optimized Task Scheduling (BCMABROTS) is developed for efficient service provisioning in the cloud environment. The BCMABROTS method begins by collecting the number of incoming user-requested tasks. After that, the resources of virtual machines, such as energy, bandwidth, memory, and CPU, are measured. The ensemble classifier constructs the set of weak learners as the nearest prototype centroid classifier. The weak classifier uses Bivariate Correlation to classify the resource-optimal virtual machine based on the available resources. The results of the weak learners are combined to provide strong classification results. Once the optimal virtual machine is categorized, the task is scheduled for that particular virtual machine by the task assigner. This approach ensures efficient cloud service provisioning with minimal time consumption. Experimental evaluation is carried out to assess factors such as task scheduling efficiency, makespan, throughput and average response time in relation to the number of cloud-requested tasks. The observed performance results confirm that the BCMABROTS method improves the task scheduling efficiency, throughput and minimizes the makespan as well as the average response time than the conventional machine learning methods.