Performance Analysis of Various Scheduling Algorithms using CloudSim

Cloud computing technologies have quickly changed how companies and organizations manage their IT resources. The core of this transformation has evolved as cloud datacenters, which offer scalable and affordable options for hosting and administering a variety of applications and services. One information technology typology that has been widely employed to deliver a range of services via the Internet is cloud computing. It guarantees simpler access to premium services and resources. Cloud systems' operation needs to be planned in order to effectively deliver services to individuals. Task scheduling seeks to maximize system throughput and distribute diverse computational resources to software programs. The unpredictability of the scenario grows as the task and has a strong potential for successful resolution. The study begins with an experimental setup to analyse the various performance metrics of task scheduling algorithms. Every experiment has several important stages. To replicate scenarios found in the real world where jobs are divided across many computing resources, the tasks are assigned to available data centers. A number of experiments were carried out to analyse the performance of First Come First Service (FCFS), Shortest Job First (SJF), Round Robin (RR) and Particle Swarm Optimization (PSO) scheduling algorithms using the parameters: makespan, average completion time, average waiting time, and average cost consumption. Thus, this study provides a description of task scheduling and the performance analysis of algorithms to task scheduling that is employed in cloud computing environments.