Extending the database relational model to capture more meaning
ACM Transactions on Database Systems (TODS)
A relational model of data for large shared data banks
Communications of the ACM
Petri-Net-Based Modeling and Evaluation of Pipelined Processing of Concurrent Database Queries
IEEE Transactions on Software Engineering
The Fifth Generation grail: A survey of related research
ACM '84 Proceedings of the 1984 annual conference of the ACM on The fifth generation challenge
| Hi-index | 0.00 |
A relational database system based on the principles of functional, data-driven computation is proposed. An architecture composed of a large number of independent, asynchronously operating processing units, each equiped with a separate memory unit holding a portion of the database is shown which implements the relational database system. All processing units are interconnected via 1) a circular shift-register bus, and 2) a daisy chain connection. Relations are represented as streams of values (tuples) where each value is carried by a unique token. To perform a query on the data base, relations involved in the query are replicated from the database as streams and used as inputs to dataflow programs (graphs) implementing the relational algebra. All operators represented within the dataflow program are executed by independent processing units and are data-driven. Proceeding asynchronously as stream values needed as inputs are produced. A time complexity of O(n) is achieved for any relational algebra query assuming a sufficient number of processing units are available.