一种数据库服务多点容灾系统  

Database service multi-point disaster tolerance system

作　　者：卢正添[1] 李涛[1] 赵奎[1] 胡晓勤[1] 刘才铭[2] 

机构地区：[1]四川大学计算机学院,成都610065 [2]中国石油西南油气田分公司通信公司,成都610051

出　　处：《计算机应用研究》2008年第12期3707-3709,共3页Application Research of Computers

基　　金：国家“863”计划资助项目(2006AA01Z435);国家自然科学基金资助项目(60573130,60502011);国家教育部博士点基金资助项目(20070610032)

摘　　要：设计并实现了一种数据库服务多点容灾系统。该系统实时监控主数据库的数据变化,将监控到的数据变化在多个异地备用数据库上进行实时重放,保证备用数据库与主数据库的数据一致性。同时对主数据库进行失效检测,及时发现故障,并以较短的时间(秒级)完成数据库服务的漂移,确保服务连续性,提高数据库服务的容灾抗毁能力。This paper designed and implemented a database service multi-point disaster tolerance system. The system monitored the data changes of the primary database in real-time. And the data changes, which were monitored, were replayed on the several remote standby databases in real-time, ensured the data consistency of the standby databases and the primary database. The system detected the failure of the primary database. When the primary database failed, the database service would be automatically migrated to a standby database in few seconds, ensured the service continuity, and improved the disaster tolerance capacity of the database service.

关 键 词：数据镜像 服务容灾 失效检测 服务漂移 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]