Performance Improvement in Cloud Computing Using Resource Clustering

Abstract

Cloud computing is a computing paradigm in which the various tasks are assigned to a combination of connections, software and services that can be accessed over the network. The computing resources and services can be efficiently delivered and utilized, making the vision of computing utility realizable. In various applications, execution of services with more number of tasks has to perform with minimum intertask communication. The applications are more likely to exhibit different patterns and levels and the distributed resources organize into various topologies for information and query dissemination. In a distributed system the resource discovery is a significant process for finding appropriate nodes. The earlier resource discovery mechanism in cloud system relies on the recent observations. In this study, resource usage distribution for a group of nodes with identical resource usages patterns are identified and kept as a cluster and is named as resource clustering approach. The resource clustering approach is modeled using CloudSim, a toolkit for modeling and simulating cloud computing environments and the evaluation improves the performance of the system in the usage of the resources. Results show that resource clusters are able to provide high accuracy for resource discovery.