Scalable Availability Model

Abstract

Guaranteeing the availability of the trust management service is a difficult problem due to the unpredictable number of cloud service consumers and the highly dynamic nature of the cloud services. For example, if the trust management service is down for a while (e.g., overload or service update), the cloud service consumers will be unable to give feedbacks or inquire a trust assessment for cloud services. Consequently, approaches that require understanding of consumers’ interests and capabilities through similarity measurements or operational availability measurements (i.e., uptime to the total time) are inappropriate in the cloud environment. The trust management service should be adaptive and highly scalable to be functional in cloud environments.