Relational Division: Four Algorithms and Their Performance

We outline three known algorithms for relational division, the algebra operator used to express universal quantification (for-all conditions), and a new algorithm called Hash-Division. By comparing the algorithms analytically and experimentally, we show that the new algorithm provides performance competitive or superior to techniques used to-date, namely techniques using sorting or aggregate functions. Furthermore, the new algorithm can eliminate duplicates in the divisor on the fly, ignores duplicates in the divided, and allows two kinds of partioning, either of which can be used to resolve hash table overflow or to efficiently implement the algorithm on a multi-processor system. Relational Division: Four Algorithms and Their Performance Goetz Graefe Oregon Graduate Center Abstract We outline three known algorithms for relational division, the algebra operator used t o express universal quantification (for-all conditions), and a new algorithm called Hash-Division. By comparing the algorithms analytically and experimentally, we show tha t the new algorithm provides performance competitive or superior t o techniques used to-date, namely techniques using sorting or aggregate functions. Furthermore, the new algorithm can eliminate duplicates in the divisor on the fly, ignores duplicates in the dividend, and allows two kinds of partitioning, either of which can be used t o resolve hash table overflow or t o efficiently implement the algorithm on a multi-processor system.We outline three known algorithms for relational division, the algebra operator used t o express universal quantification (for-all conditions), and a new algorithm called Hash-Division. By comparing the algorithms analytically and experimentally, we show tha t the new algorithm provides performance competitive or superior t o techniques used to-date, namely techniques using sorting or aggregate functions. Furthermore, the new algorithm can eliminate duplicates in the divisor on the fly, ignores duplicates in the dividend, and allows two kinds of partitioning, either of which can be used t o resolve hash table overflow or t o efficiently implement the algorithm on a multi-processor system.