Architectures for Trusted Utility Computing

Building an architecture requires a complete understanding of its building blocks and the relationships among the building blocks, because only then can it be decomposed, constructed or combined with other architectures efficiently and without conflicts. This understanding is also essential for the prediction of the properties that a given architecture will posses and hence allows the evaluation of a given architecture's fitness for a set of requirements. Within the trusted computing and security communities, a number of architectures have been proposed to solve specific problems within the utility computing domain. These architectures, including trusted virtualisation, trusted network connect, trusted compilation, are designed to specifically address a subset of the security problems within utility computing. To enable the development of sound architectures for trusted computing, we need to understand the composition of the various architectures. This knowledge allows us to answer questions such as; what are the components, connectors, properties and principles behind a given architecture, how can we combine two or more architectures to take advantage of their properties and how can we determine how trustworthy an infrastructure built using a given architecture would be?