公开(公告)号：US08103621B2

公开(公告)日：2012-01-24

申请号：US12245165

申请日：2008-10-03

申请人： Carsten Brixius ,  Dietmar Fischer ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

发明人： Carsten Brixius ,  Dietmar Fischer ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

IPC分类号： G06F17/00

CPC分类号： G06F17/30221 ,  G06F17/30153

摘要： One aspect of the present invention provides an operation for recovering lost space and maintaining data consistency in a storage environment by providing a two-way orphan reconciliation method for extremely large file systems, such as a Hierarchical Storage Management (HSM) system which manages billions of files. This highly scalable and parallelizable orphan identification process may be used in a HSM controlled environment to enable the execution of a two-way file orphan check. In one embodiment, this orphan check identifies file system client orphans and server object orphans in a single pass by comparing a queue containing a list of migrated files on a storage repository server with a queue containing a list of stub files on a file system. If the queue elements do not match, a file system orphan or server object orphan can appropriately be identified.

摘要翻译： 本发明的一个方面提供了一种用于恢复存储环境中的丢失空间并维持数据一致性的操作，其中提供用于极大文件系统的双向孤立对照方法，诸如管理数十亿个 文件。 这种高度可扩展和可并行化的孤立识别过程可以在HSM控制环境中使用，以执行双向文件孤立​​检查。 在一个实施例中，这种孤立检查通过将包含存储库服务器上的迁移文件列表的队列与包含文件系统上的存根列表列表的队列进行比较来识别文件系统客户端孤儿和服务器对象孤儿。 如果队列元素不匹配，则可以适当地识别文件系统孤立或服务器对象孤立。

公开(公告)号：US20110145196A1

公开(公告)日：2011-06-16

申请号：US13035095

申请日：2011-02-25

申请人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

发明人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

IPC分类号： G06F12/16 ,  G06F17/30

CPC分类号： G06F11/1435 ,  G06F11/1451 ,  G06F11/1458 ,  G06F11/1464 ,  G06F11/1469 ,  G06F17/30153 ,  G06F17/30221

摘要： Performing a backup of a stub object located on a file system managed by a hierarchical storage manager configured to migrate data objects from the file system to a migration storage pool. The stub object includes information for recalling a migrated data object and is configured to determine whether a backup copy of the migrated data object is stored in a backup storage pool if the backup is performed in an incremental backup operation. The backup further includes directing the hierarchical storage manager to recall the migrated data object to the file system if the backup copy of the migrated data object is not stored in the backup storage pool or if the backup is performed in a selective backup operation. The backup additionally includes storing a backup copy of the migrated data object in the backup storage pool if the migrated data object is recalled.

摘要翻译： 执行位于由分层存储管理器管理的文件系统上的存根对象的备份，配置为将数据对象从文件系统迁移到迁移存储池。 存根对象包括用于调用迁移的数据对象的信息，并且被配置为确定如果在增量备份操作中执行备份，则迁移的数据对象的备份副本是否存储在备份存储池中。 备份还包括如果迁移的数据对象的备份副本未存储在备份存储池中或者如果在选择性备份操作中执行备份，则指示分层存储管理器将迁移的数据对象调回到文件系统。 备份还包括如果迁移的数据对象被调用，则将迁移的数据对象的备份副本存储在备份存储池中。

公开(公告)号：US07272690B2

公开(公告)日：2007-09-18

申请号：US10739209

申请日：2003-12-17

申请人： David W. Groves ,  Michael L. Lamb ,  Douglas S. Noddings ,  Raymond M. Swank ,  Kevin J. Webster

发明人： David W. Groves ,  Michael L. Lamb ,  Douglas S. Noddings ,  Raymond M. Swank ,  Kevin J. Webster

IPC分类号： G06F12/00

CPC分类号： G06F3/0631 ,  G06F3/0607 ,  G06F3/067

摘要： An example of a method for assigning storage includes receiving a request for storage, wherein the request includes a requested amount of storage space and connectivity information. This example of the method also includes identifying storage that is accessible by a requesting device based on the connectivity information, and identifying all LUNs that are masked from all hosts, in the identified storage. This example of the method further includes selecting at least one identified LUN, and assigning the at least one selected LUN to the requesting device.

摘要翻译： 用于分配存储的方法的示例包括接收存储请求，其中所述请求包括请求的存储空间量和连接性信息。 该方法的该示例还包括基于连接性信息识别请求设备可访问的存储，以及识别在所识别的存储器中从所有主机屏蔽的所有LUN。 该方法的该示例还包括选择至少一个识别的LUN，以及将至少一个所选择的LUN分配给请求设备。

公开(公告)号：US20100088271A1

公开(公告)日：2010-04-08

申请号：US12245165

申请日：2008-10-03

申请人： Carsten Brixius ,  Dietmar Fischer ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

发明人： Carsten Brixius ,  Dietmar Fischer ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

IPC分类号： G06F17/30

CPC分类号： G06F17/30221 ,  G06F17/30153

摘要： One aspect of the present invention provides an operation for recovering lost space and maintaining data consistency in a storage environment by providing a two-way orphan reconciliation method for extremely large file systems, such as a Hierarchical Storage Management (HSM) system which manages billions of files. This highly scalable and parallelizable orphan identification process may be used in a HSM controlled environment to enable the execution of a two-way file orphan check. In one embodiment, this orphan check identifies file system client orphans and server object orphans in a single pass by comparing a queue containing a list of migrated files on a storage repository server with a queue containing a list of stub files on a file system. If the queue elements do not match, a file system orphan or server object orphan can appropriately be identified.

摘要翻译： 本发明的一个方面提供了一种用于恢复存储环境中的丢失空间并维持数据一致性的操作，其中提供用于极大文件系统的双向孤立对照方法，诸如管理数十亿个 文件。 这种高度可扩展和可并行化的孤立识别过程可以在HSM控制环境中使用，以执行双向文件孤立​​检查。 在一个实施例中，这种孤立检查通过将包含存储库服务器上的迁移文件列表的队列与包含文件系统上的存根列表列表的队列进行比较来识别文件系统客户端孤儿和服务器对象孤儿。 如果队列元素不匹配，则可以适当地识别文件系统孤立或服务器对象孤立。

公开(公告)号：US20090249005A1

公开(公告)日：2009-10-01

申请号：US12056389

申请日：2008-03-27

申请人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

发明人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

IPC分类号： G06F12/00

CPC分类号： G06F11/1435 ,  G06F11/1451 ,  G06F11/1458 ,  G06F11/1464 ,  G06F11/1469 ,  G06F17/30153 ,  G06F17/30221

摘要： A method for performing a backup of a stub object located on a file system managed by a hierarchical storage manager configured to migrate data objects from the file system to a migration storage pool is provided. The stub object includes information for recalling a migrated data object. The method comprises determining whether a backup copy of the migrated data object is stored in a backup storage pool if the backup is performed in an incremental backup operation; directing the hierarchical storage manager to recall the migrated data object to the file system if the backup copy of the migrated data object is not stored in the backup storage pool or if the backup is performed in a selective backup operation; creating the backup copy of the migrated data object if the migrated data object is recalled; storing the backup copy of the migrated data object in the backup storage pool if the migrated data object is recalled; creating a backup copy of the stub object if the migrated data object is not recalled; storing the backup copy of the stub object from the file system in the backup storage pool if the migrated data object is not recalled; and logically grouping the backup copy of the migrated data object with the backup copy of the stub object in the backup storage pool such that the backup copy of the migrated data object cannot be deleted from the backup storage pool unless the backup copy of the stub object does not exist in the backup storage pool if the migrated data object is not recalled.

摘要翻译： 提供了一种用于执行位于由被配置为将数据对象从文件系统迁移到迁移存储池的分层存储管理器管理的文件系统上的存根对象的备份的方法。 存根对象包括用于调用迁移的数据对象的信息。 该方法包括：如果在增量备份操作中执行备份，则确定所迁移的数据对象的备份副本是否存储在备份存储池中; 如果迁移的数据对象的备份副本未存储在备份存储池中或者如果在选择性备份操作中执行备份，则指示分级存储管理器将迁移的数据对象调回到文件系统; 如果迁移的数据对象被调用，则创建迁移的数据对象的备份副本; 如果迁移的数据对象被调用，则将迁移的数据对象的备份副本存储在备份存储池中; 如果未调用迁移的数据对象，则创建存根对象的备份副本; 如果未调用迁移的数据对象，则将存根对象的备份副本从文件系统存储在备份存储池中; 并将迁移的数据对象的备份副本与备份存储池中的存根对象的备份副本进行逻辑分组，以便不能从备份存储池中删除已迁移数据对象的备份副本，除非存根对象的备份副本 如果迁移的数据对象未被调用，则不在备份存储池中。

公开(公告)号：US08055862B2

公开(公告)日：2011-11-08

申请号：US13035095

申请日：2011-02-25

申请人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

发明人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

IPC分类号： G06F12/00

CPC分类号： G06F11/1435 ,  G06F11/1451 ,  G06F11/1458 ,  G06F11/1464 ,  G06F11/1469 ,  G06F17/30153 ,  G06F17/30221

摘要： Performing a backup of a stub object located on a file system managed by a hierarchical storage manager configured to migrate data objects from the file system to a migration storage pool. The stub object includes information for recalling a migrated data object and is configured to determine whether a backup copy of the migrated data object is stored in a backup storage pool if the backup is performed in an incremental backup operation. The backup further includes directing the hierarchical storage manager to recall the migrated data object to the file system if the backup copy of the migrated data object is not stored in the backup storage pool or if the backup is performed in a selective backup operation. The backup additionally includes storing a backup copy of the migrated data object in the backup storage pool if the migrated data object is recalled.

摘要翻译： 执行位于由分层存储管理器管理的文件系统上的存根对象的备份，配置为将数据对象从文件系统迁移到迁移存储池。 存根对象包括用于调用迁移的数据对象的信息，并且被配置为确定如果在增量备份操作中执行备份，则迁移的数据对象的备份副本是否存储在备份存储池中。 备份还包括如果迁移的数据对象的备份副本未存储在备份存储池中或者如果在选择性备份操作中执行备份，则指示分层存储管理器将迁移的数据对象调回到文件系统。 备份还包括如果迁移的数据对象被调用，则将迁移的数据对象的备份副本存储在备份存储池中。

公开(公告)号：US07953945B2

公开(公告)日：2011-05-31

申请号：US12056389

申请日：2008-03-27

申请人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

发明人： Stefan Bender ,  Dietmar Fischer ,  Douglas S. Noddings ,  James P. Smith

IPC分类号： G06F12/00

CPC分类号： G06F11/1435 ,  G06F11/1451 ,  G06F11/1458 ,  G06F11/1464 ,  G06F11/1469 ,  G06F17/30153 ,  G06F17/30221

摘要： A method for performing backup of a stub object located on a file system managed by a hierarchical storage manager configured to migrate data objects from the file system 10 a migration storage pool includes determining whether a backup copy of the migrated data object is stored in a backup storage pool if the backup is performed in an incremental backup operation. The storage manager recalls the migrated data object to the file system if the backup copy of the migrated data object is not stored in the backup storage pool or if the backup is performed in a selective backup operation. If the migrated data object is recalled, the backup copy of the migrated data object is created and stored in the backup storage pool. If the migrated data object is not recalled, a backup copy of the stub object is created and stored in the backup storage pool.

摘要翻译： 一种用于执行位于由分层存储管理器管理的文件系统上的存根对象的备份的方法，该分层存储管理器被配置为从文件系统10迁移数据对象迁移存储池包括确定迁移的数据对象的备份副本是否存储在备份中 如果在增量备份操作中执行备份，则存储池。 如果迁移的数据对象的备份副本未存储在备份存储池中，或者在选择性备份操作中执行备份，则存储管理器会将已迁移的数据对象调回到文件系统。 如果迁移的数据对象被调用，则迁移的数据对象的备份副本将被创建并存储在备份存储池中。 如果迁移的数据对象未被调用，则创建存根对象的备份副本并将其存储在备份存储池中。

公开(公告)号：US20100191708A1

公开(公告)日：2010-07-29

申请号：US12358494

申请日：2009-01-23

申请人： Carsten Brixius ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

发明人： Carsten Brixius ,  Wayne C. Hineman ,  Christian Mueller ,  Douglas S. Noddings ,  Wayne A. Sawdon

IPC分类号： G06F12/02 ,  G06F17/30 ,  G06F9/46

CPC分类号： G06F16/162

摘要： A method of synchronous deletion of managed files in a file system includes receiving a destroy event for a file to be deleted from the file system, the destroy event being generated upon request to destroy a file or corresponding objects of the files system; processing the received destroy event. Processing the destroy event includes determining if hierarchical storage management of the file system is initiated, and if initiated, continuing processing of the received destroy event; blocking threads indefinitely for an event storm during processing of the received destroy event; determining if the file to be deleted is being premigrated, migrated or is being recalled; aborting migration of the file based on the determination of migration and recall; and deleting the file and server objects corresponding to the file from the file system, where initiation of file deletion and server object deletion are synchronous.

摘要翻译： 在文件系统中同步删除被管理文件的方法包括从文件系统接收要删除的文件的销毁事件，销毁事件是根据要求销毁文件或文件系统的对应对象而产生的; 处理收到的销毁事件。 处理破坏事件包括确定是否启动文件系统的分层存储管理，并且如果启动，则继续处理所接收的销毁事件; 在处理接收到的破坏事件期间无限期阻止线程发生事件风暴; 确定要删除的文件是否被预先迁移，迁移或被召回; 根据迁移和召回的决定中止档案的移交; 并从文件系统中删除与文件相对应的文件和服务器对象，其中文件删除和服务器对象删除的启动是同步的。