Carnot and InfoSleuth

Abstract

Carnot Database TechnologyThe Carnot research project [CARN, WOEL93] at MCC was initiated in 1990 with the goal of addressing the problem of logically unifying physically-distributed, enterprisewide, heterogeneous information. A prototype has been implemented that provides services for enterprise modeling and model integration to create au enterprise-wide view, semantic expansion of queries on the view to queries on individual resources, and interresource consistency management. Carnot also includes technology for 3D visualization of large information spaecs, knowledge discovery in databases, and software application design recovery. The Camot prototype software has been used by the sponsors of the Carnot project to develop a number of applications. These applications have included worldtow management, heterogeneous database access, knowledge discovery in large databases, and integrated access to both text databases and structured databases from a single initial query. The implementation of the Carnot system has required unique advances in two technology areas. First, innovative techniques for kuowledge representation have been developed to capture and maintain an enterprise model and to map operations between an enterprise model and the physical databases. Second, a flexible, dynamic, distributed processing enviroomcnt has been developed that supports the automatic generation of program scripb that execute on heterogeneous, distrihutcd systems. The scripts control the flow of processing and can be reconfigured dynamically to respond to changes in the hardware environment or to the incorporation of additional information resources. The scripts are embedded in autonomous computing agents that can be dispatched to remote sites.