%0 Journal Article
%A FENG Xia
%A ZHANG Jiang
%A ZUO Hai-chao
%T A Dynamic Adjustment Strategy of Materialized Views Based on Query Clustering
%D 2018
%R 10.13190/j.jbupt.2017-234
%J Journal of Beijing University of Posts and Telecommunications
%P 16-22
%V 41
%N 4
%X In order to improve performance of query response of data warehouse, and avoid the frequent "jitter" phenomenon for materialized views set caused by immediate adjustment algorithm, a dynamic adjustment strategy of materialized views based on query clustering is presented. Firstly, attribute similarity can be calculated based on method of mining association rules, then queries similarity can be calculated and candidate views set can be generated by clustering the queries set during a statistical time, and then the benefits of candidate views can be calculated according to benefit model. Finally, the latest materialized views can be selected using dynamic management algorithm of materialized views. Based on the experimental results with data of air ticket settlement recorded by airlines. Whether in single-machine environment or distributed environment, compared to other benchmark algorithms, the overall performance of query response of data warehouse has been improved greatly, especially for high frequency queries.
%U https://journal.bupt.edu.cn/EN/10.13190/j.jbupt.2017-234