Data Management in Unattended Monitoring Systems

Unattended monitoring of high-value assets in cooperative environments poses significant challenges in the data management arena. For example, special procedures may need to be developed to support data authentication key management, particularly if private key authentication is used. In addition, both parties must decide what system level events should be collected to indicate the state of the monitoring system itself; what data processing, if any, should be done on jointly controlled data collection systems; and how the data will be disseminated to both parties for storage and further analysis. Both the inspecting party and the host will need sophisticated data analysis tools to compare the data to facility declarations and to resolve any discrepancies that may occur. In this paper, we describe a data management architecture under development at Sandia National Laboratories to address these issues. We identify key requirements, assumptions and design principles. Pieces of the architecture have been prototyped using the Material Monitoring System, Knowledge Generation, and commercially available components including report generation and data mirrors. We describe the current implementation and discuss lessons learned.