Interpreting remote objects at a local site
    1.
    发明授权
    Interpreting remote objects at a local site 有权
    在本地站点解释远程对象

    公开(公告)号:US08359336B2

    公开(公告)日:2013-01-22

    申请号:US10949102

    申请日:2004-09-24

    IPC分类号: G06F7/00

    CPC分类号: G06F17/30569

    摘要: A method and apparatus for receiving objects is provided. Embodiments provide for receiving, from a source database server, at a target database server, an object. If the object type of the object is involved in an inheritance hierarchy, then the target database server inspects the object to determine a source identifier that identifies to a source database server the object type associated with the object. The target database determines a target identifier that identifies to the target database server the object type associated with the object based on a mapping that associates source identifiers to target identifiers. Alternately, if the object type of the object is not involved in an inheritance hierarchy, then the target database server inspects the object to determine a set of characteristics of the object. The target database server determines the target identifier for the object based on the characteristics of the object.

    摘要翻译: 提供了一种用于接收物体的方法和装置。 实施例提供从源数据库服务器在目标数据库服务器处接收对象。 如果对象的对象类型涉及继承层次结构,则目标数据库服务器将检查对象以确定向源数据库服务器标识与对象关联的对象类型的源标识符。 目标数据库基于将源标识符与目标标识符相关联的映射来确定向目标数据库服务器标识与对象相关联的对象类型的目标标识符。 或者,如果对象的对象类型不涉及继承层次结构,则目标数据库服务器将检查对象以确定对象的一组特征。 目标数据库服务器基于对象的特征来确定对象的目标标识符。

    Replication and mapping mechanism for recreating memory durations
    2.
    发明授权
    Replication and mapping mechanism for recreating memory durations 有权
    用于重建内存持续时间的复制和映射机制

    公开(公告)号:US07752225B2

    公开(公告)日:2010-07-06

    申请号:US11873410

    申请日:2007-10-16

    IPC分类号: G06F7/00 G06F17/30

    CPC分类号: G06F17/30501 G06F9/5016

    摘要: Techniques for migrating duration trees from a source database server (or instance) to a destination database server (or instance) are provided. A duration is a memory management construct that allows database components to group related areas of memory together. A source duration tree is captured at the source database server and combined with a destination duration tree at the destination database server. Any duration identifiers in the source duration tree that conflict with (i.e., are the same as) a duration identifier in the destination duration tree are mapped to new duration identifiers.

    摘要翻译: 提供了将持续时间树从源数据库服务器(或实例)迁移到目标数据库服务器(或实例)的技术。 持续时间是一种内存管理结构,允许数据库组件将相关的内存区域组合在一起。 在源数据库服务器上捕获源持续时间树,并与目标数据库服务器上的目标持续时间树组合。 在源持续时间树中与目标持续时间树中的持续时间标识符冲突(即,与之相同)的任何持续时间标识符被映射到新的持续时间标识符。

    REPLICATION AND MAPPING MECHANISM FOR RECREATING MEMORY DURATIONS
    3.
    发明申请
    REPLICATION AND MAPPING MECHANISM FOR RECREATING MEMORY DURATIONS 有权
    用于恢复记忆时间的复制和映射机制

    公开(公告)号:US20090100082A1

    公开(公告)日:2009-04-16

    申请号:US11873410

    申请日:2007-10-16

    IPC分类号: G06F17/30

    CPC分类号: G06F17/30501 G06F9/5016

    摘要: Techniques for migrating duration trees from a source database server (or instance) to a destination database server (or instance) are provided. A duration is a memory management construct that allows database components to group related areas of memory together. A source duration tree is captured at the source database server and combined with a destination duration tree at the destination database server. Any duration identifiers in the source duration tree that conflict with (i.e., are the same as) a duration identifier in the destination duration tree are mapped to new duration identifiers.

    摘要翻译: 提供了将持续时间树从源数据库服务器(或实例)迁移到目标数据库服务器(或实例)的技术。 持续时间是一种内存管理结构,允许数据库组件将相关的内存区域组合在一起。 在源数据库服务器上捕获源持续时间树,并与目标数据库服务器上的目标持续时间树组合。 在源持续时间树中与目标持续时间树中的持续时间标识符冲突(即,与之相同)的任何持续时间标识符被映射到新的持续时间标识符。

    Reducing memory used by metadata for duplicate user defined types
    4.
    发明授权
    Reducing memory used by metadata for duplicate user defined types 有权
    减少由重复的用户定义类型的元数据使用的内存

    公开(公告)号:US07711695B2

    公开(公告)日:2010-05-04

    申请号:US11038596

    申请日:2005-01-18

    IPC分类号: G06F17/30 G06F15/16

    CPC分类号: G06F17/30595

    摘要: A computer automatically identifies multiple occurrences of an abstract data type that is embedded within an object of metadata of another abstract data type (hereinafter “top-level” abstract data type), e.g. by recursively visiting each element of the top-level abstract data type. Then the computer automatically generates, for the top-level abstract data type, a tap-level description that contains a description of each embedded abstract data type. At several locations in the top-level description that correspond to occurrences of a given embedded abstract data type in the top-level abstract data type, the computer automatically inserts offsets that identify a common position where the single description of that given embedded abstract data type is present in the tap-level description. Use of multiple offsets that identify a common position of an embedded description reduces space. The space was otherwise occupied by multiple embedded descriptions in prior art's top-level descriptions.

    摘要翻译: 计算机自动识别嵌入在另一抽象数据类型(以下称为“顶级”抽象数据类型)的元数据的对象内的抽象数据类型的多次出现。 通过递归访问顶级抽象数据类型的每个元素。 然后,计算机会自动为顶层抽象数据类型生成包含每个嵌入式抽象数据类型描述的顶级描述。 在顶层描述中的几个位置,对应于顶级抽象数据类型中给定的嵌入式抽象数据类型的出现,计算机自动插入标识共同位置的偏移,其中给定嵌入式抽象数据类型的单个描述 存在于顶级描述中。 使用标识嵌入式描述的通用位置的多个偏移可以减少空间。 在现有技术的顶级描述中,减少的空间被多个嵌入式描述所占据。

    Packaging multiple groups of read-only files of an application's components into multiple shared libraries
    5.
    发明授权
    Packaging multiple groups of read-only files of an application's components into multiple shared libraries 有权
    将应用程序组件的多个只读文件组打包成多个共享库

    公开(公告)号:US08522205B2

    公开(公告)日:2013-08-27

    申请号:US11437899

    申请日:2006-05-19

    IPC分类号: G06F9/44

    CPC分类号: G06F8/61 G06F8/71

    摘要: Multiple groups of read-only files, for use by various components of an application program in a computer, are packaged as follows. Several executable files are prepared, each executable file corresponding to a group of read-only files used by one or more components. Each executable file includes an array that has sufficient space to accommodate the group of read-only files to be packaged therein. Each array in each executable file is overwritten with data from a corresponding group of read-only files. The presence or absence of each executable file for a group of read-only files is indicated by an index that is developed during initialization of components. Retrieval of data is from the executable files, if the index indicates, on first accessing each read-only file, that the corresponding executable file is present. If the executable file is indicated by the index as being absent, then the read-only file is directly accessed.

    摘要翻译: 多组只读文件供计算机应用程序的各种组件使用,如下所示。 准备了几个可执行文件,每个可执行文件对应于一个或多个组件使用的一组只读文件。 每个可执行文件包括一个数组,该数组具有足够的空间来容纳要封装的一组只读文件。 每个可执行文件中的每个数组都被相应的只读文件组的数据覆盖。 一组只读文件的每个可执行文件的存在或不存在由组件初始化期间开发的索引指示。 数据的检索来自可执行文件,如果索引在首次访问每个只读文件时指出相应的可执行文件存在。 如果可执行文件由索引指示为不存在,则只读文件被直接访问。

    Product packaging and installation mechanism
    6.
    发明授权
    Product packaging and installation mechanism 有权
    产品包装安装机制

    公开(公告)号:US07412700B2

    公开(公告)日:2008-08-12

    申请号:US10848635

    申请日:2004-05-18

    IPC分类号: G06F9/445 G06F12/00

    摘要: Files of software to be installed in a computer are packaged as follows. A source file is prepared to contain at least a declaration of an array of sufficient space to accommodate the files to be packaged. The source file is compiled into an executable file. Thereafter, the array in the executable file is overwritten with data from each file being packaged. Depending on the embodiment, the address of where each file begins in the executable file and each file's name may be stored in the same executable file or in a different file. The stored file names are used to identify an address of a file from its name, when the named file is first accessed. In several embodiments, the operating system loads the executable file as a library that is automatically shared by multiple instances of the application (or by multiple applications), to efficiently use physical memory.

    摘要翻译: 要安装在计算机中的软件文件包装如下。 源文件准备至少包含一个足够空间的数组的声明,以容纳要打包的文件。 源文件被编译成可执行文件。 此后,可执行文件中的数组将被来自正在打包的每个文件的数据覆盖。 根据实施例,每个文件在可执行文件中开始的地址和每个文件的名称可以存储在相同的可执行文件或不同的文件中。 存储的文件名用于在首次访问命名文件时从其名称中标识文件的地址。 在几个实施例中,操作系统将可执行文件加载为由应用程序的多个实例(或多个应用程序)自动共享的库,以有效地使用物理内存。