| With the rapid development of internet and the increase of network information, a huge amount of data was produced and stored in database. The volume of database storage has scaled magnitude of orders. At the same time, accompany with the new requirement of application, though the techniques using in traditional database has been limited, the existing database can not been discarded. How to improve query performance on the basis of effectively managing these massive data in multi-database becomes the focus on current research. In massive multi-database, union query is one of the operation which cost most, therefore, it is a research work of general and practical significance to solve the problem of improving the efficiency of executing union query to satisfy the demand of performance.This paper analyzes the problem in face of union query in massive multi-database, based of the architecture of union query system, it used parsing tree to represent union query, then process global optimization and local optimization to union query on the foundation of passing tree, which improved the performance of union query effectively. Based on the query optimizing algorithms in massive multi database, the paper also chose certain strategies as incremental multi-way join, double buffer pool and parallel multi-way merge to optimize UNION query, then designed and implemented the algorithm and strategies for union query in CORBA-Based massive parallel database.At last, this paper studies the issue of union query for massive multi-database based on StarTPMonitor. Then, on the basis of the study, implemented a union query system (UQS) for massive multi-database, resolved the problem of union query in StarTPMonitor effectively, the practical tests indicate that the function characteristics and the performance characteristics of this union query system is better than the target we desired. In consequence, the algorithm is more effective in practice for union query in massive parallel data query. |
PDF Full Text Request