Generic framework for resource management in cloud computing

Abstract

Cloud Computing promises the delivery of infinite and scalable resources. These resources include tools and applications like data storage, servers, databases, networking, software etc. For the allocation of these resources to consumers, several algorithms have been introduced and implemented in research. But there are limitations of research as these algorithms do not suitably address complex scheduling problems. So this research work introduces a framework for Resource Allocation and management of services in Cloud Computing. The research proposes general architecture and detailed design for the problem of Cloud Resource Management. A solution has been formulated at Infrastructure as a Service level to schedule the resources at User Level and System Level. At User Level, an algorithm has been proposed to solve the multi-user allocation problem in Cloud scenario. The proposed framework at User Level includes Bid-shared Auction, user strategy and equilibrium analysis. In Resource Allocation every user chooses the best strategy and the amount is considered as payoff. Several metrics such as execution time, resource utility and profit have been used for the evaluation of proposed framework. At System level several metrics have been proposed for improving the response time, average resource utilization, load balancing and performance cost of the system. The proposed model has been evaluated and tested using CloudSim simulator tool. The proposed framework has also been compared with the existing Resource Management System for average resource utility and revenue maximization. newline