Day-ahead Pricing Model for Smart Cloud

Abstract

Cloud Computingais a large scaleaparallel and distributedacomputing architecture. Services like virtualized machines, computing power, storage, and software etc. are offered through Cloud Computing. On-demand/requested services are generally on the basis of ‘Pay-as-you-go’ model and Cloud service providers charge their consumers for the services they use. Consumers should have a guarantee that the services they are paying for should be delivered to them uninterrupted and this is ensured through Service Level Agreements (SLAs) between the consumers and providers. Masses around the globe associated with information technology are now realizing the importance of Cloud Computing. Major problems associated with internet can be solved through Cloud Computing. Efficient Pricing in Cloud Computing is an emerging issue in this field. Smart Cloud is the solution for this problem, Smart Cloud provides a platform to its peers which are consumers and providers. Smart Cloud uses a Time Dependent Pricing (TDP) model which calculates the price of resources on the basis of its previous consumption. TDP helps to provide a pricing technique which balances the requirements among consumers and providers. This technique satisfies the needs of both consumers and providers. The focus of this research work is to provide an efficient pricing technique along with and a scheduling policy which helps to satisfy the service provider and resource consumer. This thesis presents a Smart Cloud which constitutes a framework that can distribute Cloud resources over a communication network. In this model a Cloud Workload Management System (CWMS) has been presented which is an interface between the service consumer and the service provider. It clusters the Cloud resources using k-mean clustering algorithm. A Time Dependent Pricing (TDP) model is used to calculate the price of resources. The Cloud resource providers send resource pricing information from their records (database) to Compromised Cost-Time Based (CCTB) scheduling policy located at the CWMS. CCTB can observe and manage consumers’ resource requirements and schedules it at peak and off-peak periods. Resources can be scheduled automatically or manually by CCTB depending upon the pricing at various hours of the day thus ensuring minimum SLA violations