Using Financial Options for Pricing of IaaS Cloud Resources

Abstract

IaaS cloud resources are offered by cloud providers using various pricing schemes such as reservation, scheduled, on-demand and spot pricing. Such pricing models suffer from either wasteful payment in case of underutilization of reserved/scheduled instances or volatile prices, particularly for spot instances. This paper explores using financial options for pricing IaaS cloud resources. We advocate using options to complement existing pricing models and to provide a cheaper way to gain access to resources without reserving them in advance while avoiding prices volatility of spot instances. An option provides the cloud consumer the right, but not the obligation, to buy or sell a cloud resource at a fixed price during a predetermined period. That means that the maximum loss associated with an option is limited to the premium paid. Various options pricing methods were studied and compared. Using extensive simulation, we show the practical utility of options particularly for handling the unpredictable workload and their efficacy in terms of cost savings.