A Fairness-Aware Pricing Methodology for Revenue Enhancement in Service Cloud Infrastructure

Abstract

Infrastructure-as-a-service (IaaS) cloud providers sell their resources as different types of virtual machines (VMs) called instances. Any resource is limited in capacity, so is the resource in cloud. Pricing models can be used as a tool for cloud providers not only to cover their costs and realize a profit but also to encourage cloud tenants to use cloud resources efficiently and achieve high utilization. In this paper, we propose a new pricing methodology that encourages cloud tenants, whose requested VMs can be allocated easily and fairly, to use more cloud service by offering them lower prices, while discouraging cloud tenants, whose requested VMs are difficult to allocate, from using cloud service by charging them higher prices, while enhancing the revenue that a cloud provider receives. We perform a case study with a multiresource allocation fairness algorithm, i.e., the dominant resource sharing algorithm. Then, we further conduct case studies with two ways of price calculation: the total unit price redistribution and total revenue redistribution. Evaluation results show that the proposed pricing model with total unit price redistribution and total revenue redistribution can increase the overall revenue of cloud providers by up to 11.60% and 11.18%, respectively.