会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 3. 发明申请
    • METHOD OF INTEGRATING IN REAL TIME LARGE VOLUMES OF UPDATES IN A DATABASE
    • 在数据库中实时更新大容量积分的方法
    • WO2010052311A1
    • 2010-05-14
    • PCT/EP2009/064777
    • 2009-11-06
    • AMADEUS S.A.S.VANHOVE, NathalieDANIELLO, RudyRUSCICA, BrigitteMIRAILLES, GuyLEMAIRE, Pierre
    • VANHOVE, NathalieDANIELLO, RudyRUSCICA, BrigitteMIRAILLES, GuyLEMAIRE, Pierre
    • G06F17/30
    • G06F17/30371
    • A method of integrating large volumes of updates in a database system (200) is described. The method comprises a step of integrating the updates supplied by a data provider (220) into a master file (MF) (210) of the database system (200). The integrating step includes the steps of: receiving the updates at the master file (210); defining in the master file (210) a coherent set of data records (511 ) for each update; during the definition of the coherent set of data records (511), for each update, generating through a logistic table (360) of the database system a unique modification identifier to uniquely identify the coherent set of data records (511); receiving the modification identifier at the master file (210); at the master file (210), assigning the modification identifier to the coherent set of data records (511); committing each update by updating (515) the master file (210) with the coherent set of data records of each update and with the modification identifier assigned to the coherent set of data records (511); further obtaining from the logistic table (360) a unique commit identifier (523) for each update that has been committed by the master file (210), the commit identifier (523) being a number that reflects the order in which the committing step is completed for each update. The method also comprises a step of loading the updates into an active image (Al) (230) of the database system. The loading step includes the steps of: retrieving (533) from the master file (210) the coherent set of data record of each update; synchronizing (535) the active image (230) and the master file (210) by successively propagating into the active image (230), in the order specified by the commit identifier, each update; completing the propagation of each update into the active image (230) when the loading of each corresponding set of data records is committed (537) by the active image (230). This allows end-users of the database system to start querying immediately the propagated updates from the active image repository.
    • 描述了在数据库系统(200)中集成大量更新的方法。 该方法包括将由数据提供者(220)提供的更新集成到数据库系统(200)的主文件(MF)(210)中的步骤。 整合步骤包括以下步骤:在主文件(210)处接收更新; 在所述主文件(210)中定义用于每个更新的一组相关的数据记录(511); 在对于每个更新的相干数据记录集合(511)的定义期间,通过数据库系统的逻辑表(360)生成唯一的修改标识符以唯一地标识数据记录的相干集合(511); 在所述主文件(210)处接收所述修改标识符; 在所述主文件(210)处,将所述修改标识符分配给所述相关的数据记录集合(511); 通过用每个更新的相干的数据记录集合更新(515)主文件(210)并分配给相关的数据记录集合(511)的修改标识符来提交每个更新; 从对数表(360)进一步获得用于由主文件(210)提交的每个更新的唯一提交标识符(523),提交标识符(523)是反映提交步骤的顺序的数字 完成每次更新。 该方法还包括将更新加载到数据库系统的活动图像(A1)(230)的步骤。 加载步骤包括以下步骤:从主文件(210)检索(533)每个更新的相干的数据记录集; 每个更新都按照由提交标识符指定的顺序连续传播到活动图像(230)中来同步(535)活动图像(230)和主文件(210); 当由活动图像(230)提交每个对应的一组数据记录的加载(537)时,完成每个更新到活动图像(230)的传播。 这允许数据库系统的最终用户开始立即从活动映像存储库查询传播的更新。