Automated ``Batch" Processing for EUVE Science Data

The EUVE project at the University of California at Berkeley uses an automated process to extract the completed scientific data products from raw spacecraft telemetry. This automation has significantly reduced the human interactions required in the process, while at the same time maintaining our data delivery requirements. The system uses UNIX, IRAF, and the Web to produce the final products. A user can start a long observation choosing from a list of pre-configured processing parameters, and have the completed science products within a few days (e.g., a nine-day observation completed processing in 50 hours). If a problem arises the software will automatically notify the user who can then diagnose and fix the problems with a minimal amount of system down time. This paper will also discuss the shortcomings of this system, many of which are a direct result of its evolutionary history. This work was funded under NASA Cooperative Agreement NCC5-138.