Broker Based Qos Centric Federated Cloud Framework With Financial Options

Abstract

Cloud computing has grasped the attention of scientific community and business industry towards the provisioning of computing resources as a utility and software as a service over a network. Access to software or hardware resources are seamless to cloud consumers in a way that provider of the service could be anywhere in the world by providing services in nebulous cloud. Most of the research studies have focused on the benefits associated with single cloud provider by provisioning and managing resources at provider’s end. This has led to the problems associated with pricing and quality of the services that cloud provider guarantees in SLA to their consumers. To tackle this issue, financial options theory in cloud computing economics has been introduced in the recent past where options can be bought in advance so that these may be utilized when required. However, QoS factors that are vital for a cloud provider in order to accommodate cloud consumer trust are not yet been addressed. We propose a federated cloud resource provisioning framework based on financial options theory to address the issue of job denial during peaks by leasing spare capacity from other cloud providers as well as satisfying cost and QoS requirements. We assess cloud providers based on the services they provide and best-fit selection is made while purchasing options to meet service level objectives. Also the issue of VM placement across multiple clouds is addressed to provide a resource sharing platform where the objectives of avoiding resource rejection and efficient service level agreement management are incorporated in the proposed framework. Using CloudSim simulator, various workloads were used to test our proposed framework. Results indicate that our broker based cloud federation framework provides a win-win scenario for both cloud providers and consumers by avoiding job denial and achieving better SLA management and resource utilization.