Study On Query Optimization Of Data Warehouse Based Immune Genetic Algorithm

In the data warehouse application, the data warehouse performance question, especially the query response time of inquires, appears more and more importantly. Under data warehouse environment, the complex query of the mass data are met frequently, including the multi-tables join, gathers the computation and so on. Although the traditional relational database has carried on the optimization to each kind of data manipulation and query processing, yet it does not have full consideration the characteristic of data warehouse itself. the data warehouse query performance quality directly will affect the application of data warehouse, therefore, will completely have the necessity to conduct the research to the data warehouse performance optimization.The multi-join query optimization of data warehouse is a NP question.Main contents of this paper is to unify the immunity thought and the simple genetic algorithm, and puts forward a new immunity genetic algorithm, and makes the adaptive improvement to the question of multi-join query optimization of data warehouse,at the same time discusses the validity of immunity genetic algorithms to solve the question.Firstly, the function and architecture of data warehouse technology and OLAP have been introduced, discussed the importance of the data warehouse multi-join query optimization; After comprehensively analying the characteristics of the data warehouse system, the basic principle and difficult of query optimization and the malpractice of traditional query optimization method, the method of unifying the immunity thought and the simple genetic algorithm which can solve the data warehouse multi-join query optimization question are proposed.Secondly, after analyzing simple genetic algorithms characteristic and the flaw in detail, and unifying the immunity though which can enhance the individual multiplicity and enhance the individual fitness of the population and the simple genetic algorithms ,an immunity genetic algorithm based on the information entropy and the fitness value arrangement for the choice probability is put forward . Considered the possible degenerated phenomenon to the population in the evolution process, through withdraws the part of effective information in each generation of memory immune body to form the immunity vaccine, affects this vaccine to the immunity genetic algorithm above, enables this algorithm to have the further improvement, the effective guarantee immune body developed along the most superior direction. The validity and astringency...