A Preference-based Resource Allocation in Cloud Computing Systems

The ‘Pay-as-per-use’ model of cloud computing requires a user to rent cloud resources or services for a period of time and make payment accordingly. This paper proposes a demand-based preferential resource allocation technique that designs a market-driven auction mechanism to identify users for resource allocation based on their payment capacities and implements a payment strategy based on a buyer's service preferences. A comparison is drawn between the proposed allocation technique and the famous off-line VCG auction mechanism and results show a performance benefit in revenues to service provider, payments of cloud users besides ensuring an optimum resources use.