公开(公告)号：US07293200B2

公开(公告)日：2007-11-06

申请号：US11213630

申请日：2005-08-26

申请人： Michael Oliver Neary ,  Ashwani Wason ,  Shvetima Gulati ,  Fabrice Ferval

发明人： Michael Oliver Neary ,  Ashwani Wason ,  Shvetima Gulati ,  Fabrice Ferval

IPC分类号： G06F11/00

CPC分类号： G06F11/30 ,  G06F11/1438 ,  G06F11/1482 ,  G06F11/2028 ,  G06F11/203

摘要： Incremental single and multiprocess checkpointing and restoration is described, which is transparent in that the application program need not be modified, re-compiled, or re-linked to gain the benefits of the invention. The processes subject to checkpointing can be either single or multi-threaded. The method includes incremental page-boundary checkpointing, as well as storage checkpointing of data files associated with applications to ensure correct restoration without the need to restore files for other application programs. Incremental and full checkpoints are asynchronously merged to ensure proper operation while reducing checkpointing delay. By way of example a user-level programming library is described for loading into the address space of the application in conjunction with a loadable kernel module (LKM) or device driver used to capture and restore process state on behalf of the application. These techniques are particularly well suited for use with high-availability (HA) protection programming.

摘要翻译： 描述了增量单和多进程检查点和恢复，这是透明的，因为应用程序不需要被修改，重新编译或重新链接以获得本发明的优点。 进行检查点处理的过程可以是单线程或多线程。 该方法包括增量页面边界检查点，以及与应用程序相关联的数据文件的存储检查点，以确保正确的恢复，而无需为其他应用程序恢复文件。 递增和完整检查点被异步合并，以确保正确的操作，同时减少检查点延迟。 作为示例，用户级编程库被描述为与用于代表应用程序捕获和恢复进程状态的可加载内核模块（LKM）或设备驱动程序一起加载到应用程序的地址空间中。 这些技术特别适用于高可用性（HA）保护编程。

公开(公告)号：US07921328B1

公开(公告)日：2011-04-05

申请号：US12106166

申请日：2008-04-18

申请人： Shvetima Gulati ,  Hitesh Sharma ,  Atul R. Pandit

发明人： Shvetima Gulati ,  Hitesh Sharma ,  Atul R. Pandit

IPC分类号： G06F11/00

CPC分类号： G06F11/2074

摘要： A mirror destination storage server receives mirror update data streams from several mirror source storage servers. Data received from each mirror is cached and periodic checkpoints are queued, but the data is not committed to long-term storage at the mirror destination storage server immediately. Instead, the data remains in cache memory until a trigger event causes the cache to be flushed to a mass storage device. The trigger event is asynchronous with respect to packets of at least one of the data streams. In one embodiment, the trigger event is asynchronous with respect to packets of all of the data streams.

摘要翻译： 镜像目的地存储服务器从多个镜像源存储服务器接收镜像更新数据流。 从每个镜像收到的数据被缓存，并且定期检查点排队，但数据不会立即在镜像目标存储服务器上提交到长期存储。 相反，数据保留在高速缓存中，直到触发事件导致缓存被刷新到大容量存储设备。 触发事件相对于至少一个数据流的数据包是异步的。 在一个实施例中，相对于所有数据流的分组，触发事件是异步的。

公开(公告)号：US20060085679A1

公开(公告)日：2006-04-20

申请号：US11213630

申请日：2005-08-26

申请人： Michael Neary ,  Ashwani Wason ,  Shvetima Gulati ,  Fabrice Ferval

发明人： Michael Neary ,  Ashwani Wason ,  Shvetima Gulati ,  Fabrice Ferval

IPC分类号： G06F11/00

CPC分类号： G06F11/30 ,  G06F11/1438 ,  G06F11/1482 ,  G06F11/2028 ,  G06F11/203

摘要： Incremental single and multiprocess checkpointing and restoration is described, which is transparent in that the application program need not be modified, re-compiled, or re-linked to gain the benefits of the invention. The processes subject to checkpointing can be either single or multi-threaded. The method includes incremental page-boundary checkpointing, as well as storage checkpointing of data files associated with applications to ensure correct restoration without the need to restore files for other application programs. Incremental and full checkpoints are asynchronously merged to ensure proper operation while reducing checkpointing delay. By way of example a user-level programming library is described for loading into the address space of the application in conjunction with a loadable kernel module (LKM) or device driver used to capture and restore process state on behalf of the application. These techniques are particularly well suited for use with high-availability (HA) protection programming.

摘要翻译： 描述了增量单和多进程检查点和恢复，这是透明的，因为应用程序不需要被修改，重新编译或重新链接以获得本发明的优点。 进行检查点处理的过程可以是单线程或多线程。 该方法包括增量页面边界检查点，以及与应用程序相关联的数据文件的存储检查点，以确保正确的恢复，而无需为其他应用程序恢复文件。 递增和完整检查点被异步合并，以确保正确的操作，同时减少检查点延迟。 作为示例，用户级编程库被描述为与用于代表应用程序捕获和恢复进程状态的可加载内核模块（LKM）或设备驱动程序一起加载到应用程序的地址空间中。 这些技术特别适用于高可用性（HA）保护编程。