An Adaptive Genetic Algorithm-Based Load Balancing-Aware Task Scheduling Technique for Cloud Computing

Abstract

Task scheduling in highly elastic and dynamic processing environments such as cloud computing have become the most discussed problem among researchers. Task scheduling algorithms are responsible for the allocation of the tasks among the computing resources for their execution, and an inefficient task scheduling algorithm results in under-or over-utilization of the resources, which in turn leads to degradation of the services. Therefore, in the proposed work, load balancing is considered as an important criterion for task scheduling in a cloud computing environment as it can help in reducing the overhead in the critical decision-oriented process. In this paper, we propose an adaptive genetic algorithm-based load balancing (GALB)-aware task scheduling technique that not only results in better utilization of resources but also helps in optimizing the values of key performance indicators such as makespan, performance improvement ratio, and degree of imbalance. The concept of adaptive crossover and mutation is used in this work which results in better adaptation for the fittest individual of the current generation and prevents them from the elimination. CloudSim simulator has been used to carry out the simulations and obtained results establish that the proposed GALB algorithm performs better for all the key indicators and outperforms its peers which are taken into the consideration.