Trio: A System for Data, Uncertainty, and Lineage

Abstract

In the Trio project at Stanford, we are building a new kind of database management system: one in which data, uncertainty of the data, and data lineage are all first-class citizens. Trio is based on an extended relational model called ULDBs, and it supports a SQL-based query language called TriQL. Trio was motivated by a number of applications including scientific data management, data cleaning and integration, information extraction systems, and others. We have completed an initial working prototype of the Trio system. We will demonstrate our prototype by illustrating through two applications how uncertainty and lineage are represented in ULDBs, how TriQL operates over ULDBs both from the user and the system perspective, and in general how data, uncertainty, and lineage can work together to support interesting new functionality.