An empirical study of most fit, max-min and priority task scheduling algorithms in cloud computing

Cloud computing is envisioned to provide each IT enabled service as a utility. From hardware to software, it is committed to fulfill all kind of needs with regard to computation, storage, development etc. at cost-effective and efficient manner. The efficient delivery of services in a cloud largely depends on how it schedules the user jobs (cloudlets) over its resources. A plenty of task scheduling algorithms are available for a cloud provider to choose from. This paper selects three most versatile and fundamental task scheduling algorithms i.e. Most Fit, Max-Min and Priority algorithm to study their comparative performance. From the practical aspect, these algorithms are simulated over cloudsim 3.0.3 toolkit with variable workload circumstances. A set of metrics is also calculated and a comparative assessment of the above mentioned algorithms has been done on the basis of these metrics.