CAREER: Automatic Generation of Software Configuration Management Repositories
University Of California-Santa Cruz, Santa Cruz CA
Investigators
Abstract
CCR-0133991 CAREER: Automatic Generation of Software Configuration Management Repositories Emmet James Whitehead Twenty-five years of research and development on Configuration Management (CM) systems has bequeathed a rich legacy of data models, system architectures, and implementation approaches. This existing knowledge about CM systems is analyzed to distill out the key lessons, and then leveraged for advanced research on CM. The research advances the maturity of CM knowledge by performing a comprehensive domain analysis of CM systems, and by developing specification languages and technology to permit automatic generation of CM repositories. A domain model permits the characterization of an existing CM system as a point in a multidimensional design tradeoff space. This research describes key design spaces in sufficient detail to permit the specification of a single point in design tradeoff space. This specification can be fed to a generator that automatically generates a matching CM system. Research challenges include the creation of a repository specification language; a language for creating detailed CM system specifications that identify specific design choices, and resolve design tradeoffs; and the development of generator technology that automatically creates source code to implement a given repository specification. This auto-generation technology provides a valuable tool for exploring advanced data models for CM systems.
View original record on NSF Award Search →