Hybrid Resource Scaling for Dynamic Workload in Cloud Computing

Abstract

In the Cloud, there are enough amount of various resources to be provided for the users according to their request. To provide services the cloud resources need to be scaled up and out. Cloud resource provisioning uses two approaches. They are horizontal and vertical scaling approaches. Horizontal scaling approaches take some minutes to configure additional machines and less utilization of resources. But this approach is available. In the vertical approach, requests are not handled by adding additional virtual machines like Horizontal, resources are added on the already executing devices within a second. The objective of this work is to develop a hybrid approach by hybridizing both scaling approaches to increase utilization of resources and provide flexible resources that can satisfy user’s requests by performing in such order Vertical, horizontal approaches. The results of the study demonstrate that the proposed approach is more efficient in comparison with the existing approach. CloudSim has been used for the implementation of the developed approach. Scaling is performed by following the capacity of the machine and resources threshold value.