| Sumario: | Cloud Computing is a fast growing technology and it offers the online accessible services to the users through the internet. The input user requests are scheduled to the multiple resources in the cloud data centers. In cloud, Job Scheduling refers to send out the computing tasks of the user among the different resources according to certain rules given in cloud environments. During the scheduling process, a certain problem occurs when the more than one task assigned to a same resource simultaneously. In order to handle the multiple tasks to various resources and achieve the higher scheduling efficiency, a novel technique Multicriteria Buffalo Resource Optimized Deep Belief Neural Learnt Scheduling (MBRODBNLS) technique is introduced in cloud computing. The MBRODBNLS technique is introduced for resource efficient job schedulingwith higher efficiency and it also serves the appropriate load balancing approach. The proposed MBRODBNLS technique uses the multiple layers for learning the given input to obtain the better solution at output layer with minimum error. Firstly, user requested jobs arriveat the cloud server from various locations. After receiving the user requested jobs, cloud serveruses the multicriteria buffalo optimization technique to determine the resource efficient virtual machine. Then the cloud server submits the jobs to resource optimized virtual machine and balance the workload. This helps to balance the workload of the cloud server. Experiment is conducted on different factors such as scheduling efficiency, false positive rate, scheduling time and energy consumption. The proposed MBRODBNLS technique offers promising outcomes in terms of higher scheduling efficiency and lesser false positive as well as minimum time consumption.
|
|---|