Study Of Query Optimization In Distributed Database

In distributed database system, it becomes more difficult and complex to Distributed Query Processing because of distribution and redundancy of data distributed. Therefore, query optimization is important in distributed database system.Based on the analysis of pioneer technology of distributed query processing, a new method for distributed query processing is proposed in this paper.The model of query processing and data dictionary are designed, and the process of query is improved in the new scheme. The method can be used to reduce the quantity of data transfer through keeping parts of query results at local.User module is added into the model of query processing,so the users can choose different criterion of query optimization according to their demand.Based on different criterion of query optimization of users,different algorithms of query optimization are adopted.A parameter is added into the optimization process of SDD-1 algorithm in order to improve the ability of parallel implementation.A new algorithm of query optimization is proposed aim at making least transfer charge.This algorithm built bloom filters using joint attribute as key words. Cut off a majority of tuples that are useless for joint result through transferring bloom filters. Then form calculational results in tables. This algorithm can make more reduction by transmitting calculational result tables between different nodes. Its validity is verified by a experiment.