Research And Implementation On Query Optimization Algorithm In Distributed Database

Distributed database system is the combination of computer network and database system. Because transmission of a huge amount of data across network is involved, query processing and optimization becomes the key factor of improving query performance in distributed database. Query processing and optimization takes the advantages of sense algorithm to reduce the data of communication, and accordingly to improve the performance of response time and reduce system cost.The main work of this paper as follow:(1) Basic technologies of distributed query optimization, classes and levels of distributed query, and strategies as well as algorithms of distributed query optimization are analyzed particularly.(2) An improved query optimization algorithm of semi-join is presented bases on the research of common query optimization algorithm, which processes semi-join in sequence. And the implementation of this new algorithm is realized by two steps of creating semi-join graph and acquiring semi-join execution strategy. The experimental result shows that this query optimization algorithm evidently reduces the data of midway results and the overall cost of communication, and improves the benefit of optimization as well.(3) A multi-join particle swarm optimization algorithm is also presented in the complicated multi-join query problem in distributed database. And the implementation of this new algorithm is realized as well. The experimental result shows that this algorithm is suitable for the complicated multi-join query optimization problem, and capable of searching and choosing optimized scheme much more quickly from relation syntax tree.