In this paper, we present a policy-based approach to achieve self-deployment and self-configuration of pervasive systems and services. In our approach, we decompose applications into a set of interconnected components. With each component, we specify the autonomic attributes required to self-configure and self-manage the operations of the components in Autonomic Component Architecture (ACA) such that the overall objectives of the applications are maintained at runtime. We use the self-deployment and configuration engines of our Autonomic Computing Environment, Autonomia, to demonstrate and validate our approach. We have demonstrated how Autonomia to automatically deploy and configure the distributed computing application of Linear Equation Solver.