Cloud-based systems have the dual promise of providing on-demand service provisioning along with a "pay-as-you-use" policy. This would require the dynamic creation of a tenancy for the client by integrating existing cloud-based services on the fly. However, current cloud-based systems lack this facility, instead simply providing a catalog of services from which the client has to pick and choose. Moreover, dynamic provisioning in the cloud requires an integrated solution across the technology stack (software, platform and infrastructure) combining functional, non-functional and resource allocation requirements. Although there have been extensive research works on Web service matching, they still fall short, since they usually consider each requirement type in isolation, and cannot provide such an integrated solution. To that end, in this talk, we investigate the features needed for dynamic service provisioning on the cloud. We propose a novel User Interface - Tenant Selector - Customizer (UTC) model and approach, which enables cloud-based services to be systematically modeled and provisioned as variants of existing service tenants in the cloud. Our approach considers functional, non-functional and resource allocation requirements together. To the best of our knowledge, ours is the first such integrated approach. We illustrate our ideas using a realistic running example, and also present a proof-of-concept prototype built on IBM's Rational Software Architect modeling tool. We also present experimental results demonstrating the practical applicability of our matching algorithm, which shows dramatic performance improvements by reduction of the search space needed for matching with the help of an elimination process.
BIOGRAPHY OF SPEAKER:
back to the list of talks