Advanced computing techniques for system health management typically address problem detection and the determination of fault causality. We have designed a health management architecture that combines these techniques for problem detection and isolation with intelligent monitoring and control techniques for problem response to make humans more effective at problem management. This integration of technologies leverages a large body of advanced computing research while providing a solution that is fundamentally human-centric. Our approach is general in that the components are designed for use in a variety of applications, from single systems to complex spacecraft consisting of multiple, coupled systems. In this paper we define eight core health management components and use these components to design a health management architecture for NASA mission operations support.