公开(公告)号：US09542396B1

公开(公告)日：2017-01-10

申请号：US13728324

申请日：2012-12-27

申请人： EMC Corporation

发明人： Sitaram Pawar ,  Jean-Pierre Bono ,  Yubing Wang ,  Kumar V. K. H. Kanteti ,  Pranit Sethi

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30368 ,  G06F11/14 ,  G06F17/30144

摘要： Managing a file system to avoid unnecessary replay of a transaction log. The file system may reside on a non-volatile storage medium having an associated transaction log residing on a non-volatile storage medium, the transaction log including a plurality of log entries, and each log entry representing one or more changes to metadata of the file system. The file system may include a first data structure including a status parameter, a value of which is indicative of whether to replay the transaction log when the file system is initialized. It may be determined whether the one or more changes to metadata represented by each of the plurality of log entries have been written in place in the file system. In response to the determination, the status parameter may be set to a value indicative of the determination.

摘要翻译： 管理文件系统以避免不必要的重播事务日志。 文件系统可以驻留在具有驻留在非易失性存储介质上的相关联的事务日志的非易失性存储介质上，该事务日志包括多个日志条目，并且每个日志条目表示文件的元数据的一个或多个改变 系统。 文件系统可以包括第一数据结构，其包括状态参数，其状态参数指示当文件系统被初始化时是否重播事务日志。 可以确定对于多个日志条目中的每一个表示的对元数据的一个或多个改变是否已被写入到文件系统中。 响应于该确定，可以将状态参数设置为指示确定的值。

公开(公告)号：US09372743B1

公开(公告)日：2016-06-21

申请号：US14042326

申请日：2013-09-30

申请人： EMC Corporation

发明人： Pranit Sethi ,  Marc A. DeSouter

IPC分类号： G06F11/00 ,  G06F11/07 ,  G06F11/14

CPC分类号： G06F11/0751 ,  G06F11/0745 ,  G06F11/0793 ,  G06F11/1471 ,  G06F11/1474

摘要： A method, computer program product, and computing system for defining a transactional log file for a data storage system including defining a transactional log file for a data storage system including a data array. A plurality of IO requests for the data storage system are processed. The transactional log file is updated to include information concerning the plurality of IO requests, wherein the information includes an LGT indicator for each of the plurality of IO requests. The occurrence of a malfunction within the data storage system is sensed.

摘要翻译： 一种用于定义用于数据存储系统的事务日志文件的方法，计算机程序产品和计算系统，包括为包括数据阵列的数据存储系统定义事务日志文件。 处理数据存储系统的多个IO请求。 交易日志文件被更新为包括关于多个IO请求的信息，其中该信息包括用于多个IO请求中的每一个的LGT指示符。 感测到在数据存储系统内出现故障。

公开(公告)号：US09021303B1

公开(公告)日：2015-04-28

申请号：US13753239

申请日：2013-01-29

申请人： EMC Corporation

发明人： Marc A. DeSouter ,  Pranit Sethi ,  Morgan Clark ,  Jean-Pierre Bono ,  Sairam Veeraswamy ,  Peter C. Bixby ,  Philippe Armangau

IPC分类号： G06F11/00 ,  G06F11/14

CPC分类号： G06F11/1417 ,  G06F11/1471 ,  G06F11/1474

摘要： A dataset is recovered after a server reboot while clients access the dataset. In response to the reboot, not-yet-completed transactions in a log are parsed to create, for each of the dataset blocks modified by these active transactions, a respective block replay list of the active transactions that modify the block. Once the block replay lists have been created, clients may access specified blocks of the dataset after on-demand recovery of the specified blocks. The on-demand recovery is concurrent with a background recovery task that replays the replay lists. To accelerate log space recovery, the parsing of the log inserts each replay list into a first-in first-out queue serviced by multiple replay threads. The queue can also be used as the cache writeback queue, so that the cache index is used for lookup of the replay list and the recovery state of a given block.

摘要翻译： 服务器重新启动后，客户端访问数据集时，将恢复数据集。 响应于重新启动，解析日志中还未完成的事务，为由这些活动事务修改的每个数据集块创建修改块的活动事务的相应块重放列表。 一旦创建了块重放列表，客户端可以在指定块的按需恢复后访问数据集的指定块。 按需恢复与重放播放列表的后台恢复任务并发。 为了加速日志空间恢复，日志解析将每个重播列表插入由多个重播线程服务的先进先出队列。 队列也可以用作高速缓存回写队列，因此缓存索引用于查找重放列表和给定块的恢复状态。

公开(公告)号：US10922277B1

公开(公告)日：2021-02-16

申请号：US14753421

申请日：2015-06-29

申请人： EMC Corporation

发明人： Marc De Souter ,  Pranit Sethi

IPC分类号： G06F16/17 ,  G06F16/182

摘要： Logging changes to file system metadata using a single log hold per cached block of metadata. A metadata log including log entries is maintained in persistent storage, and each log entry in the metadata log contains metadata modified by a corresponding file system request. A single log hold is stored for each block of cached metadata. The single log hold for a given block of cached metadata indicates a file system request that modifies the metadata contained in the block, and that corresponds to a log entry located the closest, of all log entries corresponding to file system requests that modify metadata contained in the block, to a head of the metadata log.

公开(公告)号：US09418100B1

公开(公告)日：2016-08-16

申请号：US14042369

申请日：2013-09-30

申请人： EMC Corporation

发明人： Pranit Sethi ,  Marc A. DeSouter

IPC分类号： G06F17/30 ,  G06F7/00

CPC分类号： G06F17/30227

摘要： A method, computer program product, and computing system for defining a transactional log file for a data storage system including a data array. A first plurality of IO requests for the data storage system is processed. The transactional log file is updated to include information concerning the first plurality of IO requests. An IO pointer is defined to locate a specific IO request of the first plurality of IO requests within the transactional log file. File system metadata is defined on the data array for the data storage system.

摘要翻译： 一种用于为包括数据阵列的数据存储系统定义事务日志文件的方法，计算机程序产品和计算系统。 处理数据存储系统的第一多个IO请求。 交易日志文件被更新以包括关于第一多个IO请求的信息。 定义IO指针以定位事务日志文件中的第一多个IO请求的特定IO请求。 文件系统元数据在数据存储系统的数据阵列上定义。

公开(公告)号：US09355121B1

公开(公告)日：2016-05-31

申请号：US13930506

申请日：2013-06-28

申请人： EMC Corporation

发明人： Michael Scheer ,  Morgan Clark ,  Ahsan Rashid ,  Srinivasa R. Vempati ,  Marc DeSouter ,  Pranit Sethi ,  Maher Kachmar

IPC分类号： G06F17/30

CPC分类号： G06F17/30221

摘要： A technique for managing a file system includes dividing the addressable space of a file system into multiple subspaces: a data subspace for user data; an inode subspace for inodes describing the user data; and an MDB (metadata block) subspace for indirect blocks and/or directory structures pertaining to the user data. In response to storage requests to write user files to the file system, the file system stores user data of the files in the data subspace, stores inodes pertaining to the user files in the inode subspace, and stores indirect blocks and/or directories in the MDB subspace.

摘要翻译： 用于管理文件系统的技术包括将文件系统的可寻址空间划分成多个子空间：用于数据的数据子空间; 用于描述用户数据的inode的inode子空间; 和用于与用户数据有关的间接块和/或目录结构的MDB（元数据块）子空间。 响应于将用户文件写入文件系统的存储请求，文件系统将文件的用户数据存储在数据子空间中，将与用户文件有关的inode存储在索引节点子空间中，并将间接块和/或目录存储在 MDB子空间。