Convergence-Based Task Scheduling Techniques in Cloud Computing: A Review

Abstract

The cloud computing promises various benefits that are striking to establishments and consumers of their services. These benefits encourage more business establishments, institutes, and users in need of computing resources to move to the cloud because of efficient task scheduling. Task scheduling is a means by which the tasks or job specified by users are mapped to the resources that execute them. Task scheduling problems in cloud, has been considered as a hard Nondeterministic Polynomial time (Np-hard) optimization problem. Task Scheduling is use to map the task to the available cloud resources like server, CPU memory, storage, and bandwidth for better utilization of resource in cloud. Some of the problems in the task scheduling include load-balancing, low convergence issues, makespan, etc. Convergence in task scheduling signifies a point in the search space that optimize an objective function. The non-independent tasks has been scheduled based on some parameters which includes makespan, response time, throughput and cost. In this paper, an extensive review on existing convergence based task scheduling techniques was carried out spanning through 2015 to 2019. This review would provide clarity on the current trends in task scheduling techniques based on convergence issues and the problem solved. It is intended to contribute to the prevailing body of research and will assist the researchers to gain more knowledge on task scheduling in cloud based on convergence issues.