摘要
物化视图能够提高数据仓库的查询响应能力,但物化视图集的选择却很复杂。该文提出了一个如何选择物化视图集的优化遗传算法,在有限的存储空间下,使查询性能和视图维护代价较小。提出一个查询维护代价模型,并优化了交叉算子。实验结果表明,该算法在执行代价上优于经典的遗传算法。
Materialized views can improve query response capability of data warehouse, but choosing materialized views is very complex. In this paper, an improved genetic algorithms is put forward. In the limited storage space, query performance and view maintenance costs are smaller. We propose a query and maintenance cost model, and optimize the crossover. Experimental results show that the algorithm is superior to classical genetic algorithm in the implementation price.
出处
《电脑知识与技术》
2016年第1X期203-206,共4页
Computer Knowledge and Technology
关键词
遗传算法
物化视图
代价模型
交叉算子
Genetic algorithm
materialized views
cost model
crossover
