Cloud computing is the assemblage of computing resources which are conveyed as a facility to the client or multiple tenants over the internet. Job scheduling is an indispensable and utmost vital part in any cloud environs. With growing digits of customers, scheduling becomes a determined task. Identifying the best Job scheduling method is a significant challenge to improve scheduling efficiency and minimizes the makespan in big data analytics. Task classification based on certain parameters using Ensemble classifier and firefly optimized Scheduling (FOSC) technique is introduced for scheduling large quantity of jobs to ideal virtual machine with least possible time. The classification and FOSC technique maximize the resource utilization rate across the cloud server while handling massive amount of tasks. Task classifier categorizes the tasks using Ensemble classifier based on priority. The priority level of the task is calculated based on certain parameters like task size, bandwidth and memory expectation of the task and assign to different data centers. In the Data center, the tasks are stored in one or more queue and then selection of optimal virtual machines is performed. Followed by, the tasks get scheduled using firefly optimized Round Robin Scheduling algorithm. The FOSC efficiently identify the resource optimized virtual machine and allocate the task. This helps to maximize scheduling efficiency of the cloud server.

                Index Terms: Job Scheduling, Task classification, Ensemble classifier, firefly optimized scheduling, Round Robin