A Multi-Resolution Relational Data Model
Robert L. Read, Donald S. Fussell, Abraham Silberschatz · 1992
The use of data at different levels of information content is essential to the performance of multimedia, scientific, and other large databases because it can significantly decrease I/O and communication costs. The performance advantages of such a multi-resolution scheme can only be fully exploited by a data model that supports the convenient retrieval of data at different levels of information content. In this paper we extend the relational data model to support multiresolution data retrieval. In particular, we introduce a new partial set construct, called the sandbag, that can support multi-resolution for the types of data used in a wide variety of next-generation database applications, as well as traditional applications. We extend the relational algebra operators to analogous operators on sandbags. The resulting extension of the relational algebra is sound and forms a foundation for future database management systems that support these types of next-generation applications.