Making money with clouds: Revenue optimization through automated policy decisions


Many existing IT applications exhibit strongly varying demand patterns for resources.

Accommodating an ever increasing and highly fluctuating demand requires continuous availability of

sufficient resources. To achieve this state at reasonably costs, a high degree of flexibility with respect

to the given IT infrastructure is necessary. Facing this challenge the idea of Cloud computing has

been gaining interest. In so-called Clouds resources such as CPU, storage and bandwidth can be

bundled into a single services, which are offered to Cloud users. These services can be accessed in

oblivion of the underlying IT infrastructure. This way Cloud Computing facilitates the introduction of

new products and services without large investments in the IT infrastructure.

Cloud Computing is a promising approach with a high impact on business models. One aspect of

business models is clearly the revenue model, which defines how prices should be set to achieve

predefined revenue level. The decision about accepting or denying requests has a high impact on the

revenue of the provider. In this paper we analyze two approaches that support the cloud provider in its

decision. We show that predefined policies allow increasing revenue compared to widely used

technical models such as first-come-first-serve.

This document is currently not available here.