Multivalent Optimizer-Based Hybrid Genetic Algorithm for Task Scheduling in Cloud Applications

Cloud computing platforms provide on-demand online services without the need for direct user management. Generally, big clouds distribute functions over multiple data centers at distant locations. The major facilities offered by clouds are based upon virtual machines which provide benefits in terms of low scheduling cost, improved accessibility and availability of cloud services. While transferring the tasks for effective scheduling, the main issue arises due to the domain and characteristics difference of the source machine and the target machine. During network traffic, the challenges are more complex thereby resulting in slow data transfer which leads further issues such as delayed delivery of critical tasks. In order to address the problem of heterogeneity in cloud task, there is a strong requirement of optimal scheme for task scheduling. This research work implements an optimization scheme for scheduling tasks in cloud domains. The offered scheme uses a multivalent optimizer using genetic algorithm termed as Multivalent Optimizer based Genetic Algorithm (MO-GA). It attempts to enhance system performance by transferring the tasks through cloud networks on the basis of resources workload. Therefore, it’s very important to apply proper transfer mechanisms for efficient task scheduling in cloud applications. The suggested scheme (MO-GA) ponders various parameters such as system throughput, amount of virtual machines, total number of tasks, speed and capacity. From analytical results, it can be easily identified that our scheme optimizes task scheduling even for large number of tasks efficiently. MO-GA succeeds to achieve optimized tasks’ transfer time and get promising results. The scheme is investigated using MATLAB distrusted system for the simulation of the cloud environment. The proposed scheme manage enhancement and optimization of almost 15% over the existing schemes for task transfer.