标题：Multiple sparse tables based on pivot table for multi-tenant data storage in SaaS
作者：Wang X.; Li Q.; Kong L.
作者机构：[Wang, X] School of Computer Science and Technology, University of Shandong, Jinan, Shandong Province, China;[ Li, Q] School of Computer Science and T 更多
通讯作者地址：Wang, X.; School of Computer Science and Technology, University of Shandong, Jinan, Shandong Province, China; 电子邮件: firstname.lastname@example.org
来源：2011 IEEE International Conference on Information and Automation, ICIA 2011
关键词：index; multi-tenant; pivot table; saas; sparse table
摘要：In order to excellently support SaaS application, multi-tenant database system needs to meet the tenants' requirement of isolation and on-demand customization, and then needs to provide data storage mechanism and index mechanism that supporting isolation and flexibility. Multiple Sparse Tables is a good Approach for Multi-tenant Data Storage in SaaS, but no use of physical index provided by RDBMS. Based on the Multiple Sparse Tables approach, in this paper, we proposed a meta-data driven indexing mechanism. According to tenants' customization requirement, the model constructs respective index metadata for business data of the tenants, and achieves isolation of index data & customization; meanwhile the index maintenance strategies are given. According to tenants' access requests and the Pivot Table, the model returns the tenants' result sets more quickly or updates index data on-demand. Detailed experimental results show that the index maintenance and data access of this approach works with good performance under balanced conditions. © 2011 IEEE.