Query Optimization Research And Application On Oil Well Construction Design

Query optimization is not only an important technology used in the design and realization of database management system, but also an important factor that affects the performance of database system. There are two methods of database query optimization: cost-based and response time-based query optimization, all of which can improve the efficiency of complex query. Through the study of combination these two methods, the series of the algorithms and methods to achieve to two-stage query optimize is unfolded, which is applied to the research and development of the actual project, "Oil Well Construction Design System", to resolve the problem of system response time.Cost-based query optimization is the primary technology of optimization phase one. The main objective is to equivalent transform user's query to the most efficient query plan. In order to optimize the parallel implementation phase of the multi-thread query, query plan need to be translated into standard expression and then to be further optimized, at last, the optimized implementation plan which is suitable for multi-thread parallel implementation comes out.Multi-thread parallel query technology is used in phase two in order to apply response time-based query optimization. In this paper, the multi-thread parallel optimized algorithms for big table query, big table join, aggregation query within group and sort query are unfolded, and the implementation of parallelization algorithm based on split big table method, the ways and principles to split big table and the algorithm of maintaining the data of big table are discussed.Since the system is put into operation in good shape, users receive satisfactory results, and the query efficiency is substantial increased. Through analysis and test of the query efficiency, it is figured out the system response time can be improved more than 30 times without many cost of data maintenance operation. Throughout research, this paper puts the method and algorithms of parallel computing into practical application based on normal PC server, expands the scope of parallel computing, brings performance improvement to the PC users and businesses by parallel computing, and has practical significance.