Abstract

The management of today’s very large databases requires high storage capacity and parallel searching to ensure real-time operation. The two main characteristics of database processing are: a) most of the transactions are based on comparisons of the data fields with a search argument, and b) a large amount of data have to be retrieved and processed in order to generate the query result, which is usually a small fraction of the database. Therefore, a common solution to the problem is a special purpose processor operating as a data filter. This unit receives records from the database and screens them on the fly accordingto user defined criteria. If a record satisfies the query, it is transferred to the end user, otherwise, it is rejected.

© 1993 Optical Society of America