发明名称 DATA REPLICATION IN SITE RECOVERY ENVIRONMENT
摘要 Techniques disclosed herein relate to synchronizing a first database with a second database. Embodiments include detecting a write operation modifying properties of a data object in the first database. While the data object is locked, embodiments write object change data to a journal table. Embodiments query the journal table of the recovery database to retrieve a portion of the object change data corresponding to a first window of time and comprising a plurality of entries. The retrieved portion of object change data is processed to create processed object data by collapsing duplicate entries within the plurality of entries. Embodiments retrieve object data from the first database, corresponding to properties of data objects specified in the processed object change data. The retrieved object data is pushed to the second database, whereby the second database is synchronized with the first database.
申请公布号 US2015379038(A1) 申请公布日期 2015.12.31
申请号 US201414315091 申请日期 2014.06.25
申请人 VMware, Inc. 发明人 Nikolov Dian
分类号 G06F17/30;G06F11/14 主分类号 G06F17/30
代理机构 代理人
主权项 1. A method of synchronizing a first database of a site recovery platform with a second database of a site recovery platform by pushing changes from the first database to the second database, comprising: detecting a write operation modifying one or more properties of a data object in the first database of the site recovery platform; while the data object within the first database is locked as a result of the detected write operation, writing, using a first thread of a plurality of threads, object change data based on the write operation to a journal table of the first database; querying, using a second one or more threads distinct from the plurality of threads, the journal table of the recovery database to retrieve a portion of the object change data corresponding to a first window of time, the portion comprising a plurality of entries; processing, using the second one or more threads, the retrieved portion of object change data to create processed object data by collapsing duplicate entries within the plurality of entries; retrieving object data from the first database, corresponding to properties of data objects specified in the processed object change data; and pushing the retrieved object data to the second database, whereby the second database is synchronized with the first database.
地址 Palo Alto CA US
您可能感兴趣的专利