标题：A Rapid Aggregation Computing Approach Based on Dimension Storage
作者：Song, Ai-mei; Zhao, Mao-xian; Ma, Xu
作者机构：[Song, Ai-mei; Zhao, Mao-xian] Shandong Univ Sci & Technol, Coll Informat Sci & Engn, Qingdao, Peoples R China.; [Ma, Xu] Southeast Univ, Sch Comp S 更多
会议名称：International Symposium on Signal Processing Biomedical Engineering, and Informatics (SPBEI)
会议日期：DEC 16-18, 2013
来源：INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING BIOMEDICAL ENGINEERING, AND INFORMATICS (SPBEI 2013)
摘要：The OLAP queries are ad hoc complex aggregation queries on the massive data set. Aggregation usually involves multiple columns in tables. For OLAP queries, we have to scan the whole table in relational tables to select partial data; therefore, it leads to query performance degradation. To solve this problem, this paper proposes a rapid aggregation approach based on dimension storage. Our method store data from different dimensions in different files to make the least data access, meanwhile, we adopt B+ tree as index for dimension storage. So our method improves aggregation computing performance dramatically. We demonstrate the aggregation on TPC-H benchmark. The results show that our given approach significantly boosts the performance of aggregation.