Research Area:  Cloud Computing
Scheduling plays a vital role in enhancing the resource utilization and performance in the Cloud environment. The Broker Policy is responsible for Cloudlets scheduling that decides the order in which the Cloudlets are assigned to Virtual Machines. Traditional Data Center Broker Policy executes the Cloudlets in First Come First Serve (FCFS). Shortest Job First (SJF), as its name justifies, executes the Cloudlets in the increasing order of their length or execution time. However, neither FCFS nor SJF supports the concept of priority. Nevertheless, priority scheduling is typical in Real-Time scenarios where the priority plays a vital role in deciding the schedule for execution of tasks. This paper proposes a priority-based Shortest Job First Policy for Data Center Broker in Cloud environments termed as Priority_DCB that selects processes for execution in non-decreasing order of priority. If two or more Cloudlets have the same priority, their execution order will be the Shortest Job First. The proposed approach provides a simple and easy solution for executing the Cloudlets according to their priorities. CloudSim is used to implement the proposed approach and carrying out the simulations. Results show that the proposed approach is better than the other state-of-the-art Data Center Broker Policy in terms of Cloudlets turnaround time and waiting time.
Author(s) Name:  Nitin Kumar MishraPuneet HimthaniGhanshyam Prasad Dubey
Conferrence name:  Proceedings of International Conference on Communication and Computational Technologies
Publisher name:  Springer
Volume Information:  pp 279-290
Paper Link:   https://link.springer.com/chapter/10.1007/978-981-16-3246-4_23