Distributed Database Query Optimization Algorithm To Achieve

In a distributed database system, performance of system affected by the query strategy, and a query processing cost of communication is usually caused by the amount of information to decide. Half-connection method can reduce the amount of communication data. Distributed database have the ability to handle sub queries with parallel. This paper will explore two aspects of optimization measures: minimize response time and minimize total time. Distributed query processing, some scholars have done a lot of research, but these algorithms in some respects are some limitations. As following: To a certain extent, by the impact of transmission media, data and information in transit will be subject to obstacles, such a delay would slow down the query execution process. Computer equipment constantly updated every day, high-performance devices emerging, multi-processor in the network environment has been widely used, it should take full advantage of the opportunity provided by parallel processing, so a transfer can be increased the amount of data, thereby to speed up query response speed. Some algorithms lose some function in order to achieve a number of other properties, to reduce the time and space complexity, full of abuse of the means of dynamic random access to exchange the sequence of operations.By reading a lot of books and encountered some problems and solve practical problems during my part-time job, this paper mainly finish some of the work as following:1.In this paper, a study of the previously proposed algorithm ordinary semi-connected, semi-join algorithm based on this deficiency, a half-connection based on the improved query optimization algorithm - marked the second half of the connection based on optimal algorithm of query. The algorithm is consisted of the second half and connecting two steps marked fields. Experiments show that the query optimization algorithm significantly reduced the amount of intermediate results of data; effectively reducing the overall network communication costs, improve efficiency of query.2.Data for the distributed database of information fragmentation and the distribution of the different sites on the query of the question of the research, by researching the related to the distributed database technology, the paper proposed dynamic allocation, replication algorithm, and provided logic implementation of the algorithm. Experiments show that the algorithm suitable for complex multi-site query optimization problem, to more quickly get the minimum cost of a copy of the operation, a copy of dynamically created to get the optimal solution.3.To consider the distributed data characteristics, different data at different sites, and end-users of each site accessed data object to be different, in order to arrive at the user's purpose more quickly, the query based on table structure design optimization heuristic algorithm to some extent, reduce the amount of data of communication and speed up the query speed.Finally, these algorithms on the proposed by horizontal comparison experiments to verify the results, which show that the proposed algorithm reduces some of the relationship ,and the little time ,cost, compared with traditional heuristic algorithm.