Analysis of CPU Scheduling Algorithms for Cloud Computing

TLDR: A mechanism to implement a large number of scheduling algorithms and then compute their average waiting times and average turnaround times for a common test case is more efficient one for cloud environment.

Abstract: Cloud computing refers to the advancement of distributed computing which takes the computational aspects of data processing into high-power centralized data centers over networks. It refers to the use of a centralized pool of resources which are allocated to a large number of customers on a pay-as-you-go model. This creates the need of scheduling algorithms which allow us to decide which job process, amongst the ones received will be allocated resources first for execution. Job scheduling is one of the major areas of research nowadays in the field of cloud computing as it helps not only in proper utilization of resources but also in avoidance of deadlock within the cloud infrastructure. There exist a large number of scheduling algorithms such as First Come First Serve Scheduling Algorithm, Generalized Priority Algorithm, Round Robin Scheduling Algorithm and Least Slack Time Scheduling Algorithm. This paper proposes a mechanism to implement each of these and then compute their average waiting times and average turnaround times for a common test case is more efficient one for cloud environment.