Multivariate Piecewise Regressive African Buffalo Optimization-Based Resource-Aware Task Scheduling in Cloud

Abstract

Distributed cloud computing handles a large number of tasks and provides many dynamic virtualized resources that aim to share as a service through the internet. While handling a large volume of tasks, task execution times, throughput, and makespan are the most significant metrics in practical scenarios. So, the scheduling task is essential to achieve accuracy and correctness on task completion. A novel technique called Multivariate Piecewise Regressive African Buffalo Optimization-based Resource Aware Task Scheduling (MPRABO-RATS) is introduced for improving the task scheduling efficiency and minimizing time consumption. First, the cloud user dynamically generates numerous heterogeneous tasks in the cloud environments. After receiving the tasks, the task scheduler in the cloud server finds the resourceoptimized virtual machine using the Multivariate Piecewise Regressive African Buffalo Optimization technique. The proposed optimization technique uses the Multivariate Piecewise Regression function for analyzing the different resources availablity such as CPU Time, Memory, Bandwidth, and Energy before the task scheduling. Initially, the population of the virtual machine is defined. After that, the fitness is measured using Multivariate Piecewise Regression. Based on the fitness estimation, the resource-efficient virtual machine is determined. Finally, the task scheduler assigns the tasks to the resource-optimized virtual machine with higher efficiency. Experimental evaluation is carried out in the CloudSim simulator on the factors such as task scheduling Efficiency, Throughput, Makespan, and Memory Consumption with respect to a number of tasks. The observed results indicate that the MPRABO-RATS technique offers an efficient solution in terms of achieving higher task scheduling Efficiency, Throughput, and Minimizing the Makespan as well as Memory Consumption than the conventional scheduling techniques.