The paper proposes a multi-paradigm modeling approach which aims to derive performance models from software models of enterprise SOA systems, in order to evaluate their run-time performance from the early development phases. This helps to choose between different architecture, design, and configuration alternatives in order to meet the performance requirements. The source models of the proposed transformation are the platform independent model of the service oriented system, the deployment diagram and the platform aspect models. The Performance completion feature model describes the variability in the service platform that may affect the system’s performance. It provides the choice to select between multiple platform alternatives based on the business requirements. Aspect-oriented modeling techniques are used to produce a platform-dependent model by weaving aspect models of platform operations into the platform-independent model. The platform-dependent model thus obtained represents the source models for the transformation into a performance model. The UML profile MARTE is used to annotate models with performance information.