公开(公告)号：US07617216B2

公开(公告)日：2009-11-10

申请号：US11220780

申请日：2005-09-07

Applicant: Stephen A. Fridella ,  Xiaoye Jiang ,  Sorin Faibish ,  John Forecast

Inventor： Stephen A. Fridella ,  Xiaoye Jiang ,  Sorin Faibish ,  John Forecast

IPC: G06F17/30

CPC classification number: G06F17/30194 ,  G06F17/30082

Abstract: A file server system has a cluster of server computers that share access to a file system in shared storage. One of the server computers has primary responsibility for management of access to the file system. In order to reduce the possibility of primary server overload when a large number of the clients happen to concurrently access the same file system, most metadata processing operations are offloaded to secondary server computers. This also facilitates recovery from failure of a primary server computer since only a fraction of the ongoing metadata operations of a primary server computer is interrupted by a failure of the primary server computer. For example, a secondary data mover may truncate, delete, create, or rename a file in response to a client request.

Abstract translation: 文件服务器系统具有共享对共享存储器中的文件系统的访问的服务器计算机集群。 其中一台服务器计算机主要负责管理对文件系统的访问。 为了减少当大量客户端并发访问同一文件系统时主服务器过载的可能性，大多数元数据处理操作都被卸载到辅助服务器计算机。 这还有助于从主服务器计算机的故障恢复，因为主服务器计算机的正在进行的元数据操作的一小部分被主服务器计算机的故障中断。 例如，辅助数据移动器可以截断，删除，创建或重命名文件以响应客户端请求。

公开(公告)号：US20080005468A1

公开(公告)日：2008-01-03

申请号：US11382233

申请日：2006-05-08

Applicant: Sorin Faibish ,  Stephen Fridella ,  Uday Gupta ,  Xiaoye Jiang ,  Peter Bixby

Inventor： Sorin Faibish ,  Stephen Fridella ,  Uday Gupta ,  Xiaoye Jiang ,  Peter Bixby

IPC: G06F12/16 ,  G06F12/00

CPC classification number: G06F12/0866 ,  G06F3/0622 ,  G06F3/064 ,  G06F3/0643 ,  G06F3/0665 ,  G06F3/067 ,  G06F11/1435 ,  G06F2201/84

Abstract: A cached disk array includes a disk storage array, a global cache memory, disk directors coupling the cache memory to the disk storage array, and front-end directors for linking host computers to the cache memory. The front-end directors service storage access requests from the host computers, and the disk directors stage requested data from the disk storage array to the cache memory and write new data to the disk storage. At least one of the front-end directors or disk directors is programmed for block resolution of virtual logical units of the disk storage, and for obtaining, from a storage allocation server, space allocation and mapping information for pre-allocated blocks of the disk storage, and for returning to the storage allocation server requests to commit the pre-allocated blocks of storage once data is first written to the pre-allocated blocks of storage.

Abstract translation: 缓存的磁盘阵列包括磁盘存储阵列，全局高速缓冲存储器，将高速缓存存储器耦合到磁盘存储阵列的磁盘引导器以及用于将主计算机连接到高速缓冲存储器的前端引导器。 前端总监从主机服务存储访问请求，并且磁盘控制器将请求的数据从磁盘存储阵列请求到高速缓冲存储器，并将新数据写入磁盘存储器。 前端指导器或磁盘引导器中的至少一个被编程用于磁盘存储器的虚拟逻辑单元的块分辨率，并且用于从存储分配服务器获得用于磁盘存储器的预分配块的空间分配和映射信息 并且一旦数据被首先写入到预先分配的存储块中，则返回到存储分配服务器请求提交预先分配的存储块。

公开(公告)号：US07315926B2

公开(公告)日：2008-01-01

申请号：US10945654

申请日：2004-09-21

Applicant: Stephen A. Fridella ,  Xiaoye Jiang ,  Uday K. Gupta ,  Sorin Faibish

Inventor： Stephen A. Fridella ,  Xiaoye Jiang ,  Uday K. Gupta ,  Sorin Faibish

IPC: G06F13/00 ,  G06F12/00

CPC classification number: G06F3/0659 ,  G06F3/061 ,  G06F3/0614 ,  G06F3/0622 ,  G06F3/0637 ,  G06F3/0683 ,  G06F9/526 ,  G06F2209/523

Abstract: A protocol is provided for allocating file locking tasks between primary and secondary data mover computers in a network file server. When there is frequent read access and infrequent write access to a file, a primary data mover grants read locks to the entire file to secondary data movers, and the secondary data movers grant read locks to clients requesting read access. When write access to the file is needed, the read locks to the entire file are released and the read locks granted to the clients are released or expire or are demoted to non-conflicting byte range locks managed by the primary data mover. Concurrent read and write access to the same file is then managed by the primary data mover.

Abstract translation: 提供了一种协议，用于在网络文件服务器中的主要和次要数据移动计算机之间分配文件锁定任务。 当对文件进行频繁的读取访问和不频繁的写入访问时，主数据移动器将授予对整个文件的读取锁定到辅助数据移动器，次要数据移动器将向读取访问权限的客户端授予读取锁定。 当需要对文件进行写入访问时，释放对整个文件的读取锁定，授予客户端的读取锁定将被释放或过期或降级到由主数据移动器管理的非冲突字节范围锁定。 然后由主数据移动器管理对同一文件的并发读写访问。

公开(公告)号：US20070260830A1

公开(公告)日：2007-11-08

申请号：US11382193

申请日：2006-05-08

Applicant: Sorin Faibish ,  Stephen Fridella ,  Uday Gupta ,  Xiaoye Jiang

Inventor： Sorin Faibish ,  Stephen Fridella ,  Uday Gupta ,  Xiaoye Jiang

IPC: G06F12/16

CPC classification number: G06F3/067 ,  G06F3/061 ,  G06F3/0631 ,  G06F3/064

Abstract: A primary processor manages metadata of a production dataset and a snapshot copy, while a secondary processor provides concurrent read-write access to the primary dataset. The secondary processor determines when a first write is being made to a data block of the production dataset, and in this case sends a metadata change request to the primary data processor. The primary data processor commits the metadata change to the production dataset and maintains the snapshot copy while the secondary data processor continues to service other read-write requests. The secondary processor logs metadata changes so that the secondary processor may return a “write completed” message before the primary processor commits the metadata change. The primary data processor pre-allocates data storage blocks in such a way that the “write anywhere” method does not result in a gradual degradation in I/O performance.

Abstract translation: 主处理器管理生产数据集和快照副本的元数据，而辅助处理器提供对主数据集的并发读写访问。 第二处理器确定何时对生产数据集的数据块进行第一次写入，并且在这种情况下向主数据处理器发送元数据改变请求。 主数据处理器将元数据更改提交给生产数据集，并在辅助数据处理器继续服务其他读写请求时维护快照副本。 次要处理器记录元数据更改，以便辅助处理器可能在主处理器提交元数据更改之前返回“写入完成”消息。 主数据处理器预先分配数据存储块，使得“在任何地方写入”方法不会导致I / O性能逐渐降级。

公开(公告)号：US20070055702A1

公开(公告)日：2007-03-08

申请号：US11220780

申请日：2005-09-07

Applicant: Stephen Fridella ,  Xiaoye Jiang ,  Sorin Faibish ,  John Forecast

Inventor： Stephen Fridella ,  Xiaoye Jiang ,  Sorin Faibish ,  John Forecast

IPC: G06F17/30

CPC classification number: G06F17/30194 ,  G06F17/30082

Abstract: A file server system has a cluster of server computers that share access to a file system in shared storage. One of the server computers has primary responsibility for management of access to the file system. In order to reduce the possibility of primary server overload when a large number of the clients happen to concurrently access the same file system, most metadata processing operations are offloaded to secondary server computers. This also facilitates recovery from failure of a primary server computer since only a fraction of the ongoing metadata operations of a primary server computer is interrupted by a failure of the primary server computer. For example, a secondary data mover may truncate, delete, create, or rename a file in response to a client request.

Abstract translation: 文件服务器系统具有共享对共享存储器中的文件系统的访问的服务器计算机集群。 其中一台服务器计算机主要负责管理对文件系统的访问。 为了减少当大量客户端并发访问同一文件系统时主服务器过载的可能性，大多数元数据处理操作都被卸载到辅助服务器计算机。 这还有助于从主服务器计算机的故障恢复，因为主服务器计算机的正在进行的元数据操作的一小部分被主服务器计算机的故障中断。 例如，辅助数据移动器可以截断，删除，创建或重命名文件以响应客户端请求。

公开(公告)号：US20060230148A1

公开(公告)日：2006-10-12

申请号：US11099912

申请日：2005-04-06

Applicant: John Forecast ,  Stephen Fridella ,  Sorin Faibish ,  Xiaoye Jiang ,  Uday Gupta

Inventor： John Forecast ,  Stephen Fridella ,  Sorin Faibish ,  Xiaoye Jiang ,  Uday Gupta

IPC: G06F15/173

CPC classification number: H04L69/16 ,  H04L69/14 ,  H04L69/165

Abstract: For each high-level protocol, a respective mesh of Transmission Control Protocol (TCP) connections is set up for a cluster of server computers for the forwarding of client requests. Each mesh has a respective pair of TCP connections in opposite directions between each pair of server computers in the cluster. The high-level protocols, for example, include the Network File System (NFS) protocol, and the Common Internet File System (CIFS) protocol. Each mesh can be shared among multiple clients because there is no need for maintenance of separate TCP connection state for each client. The server computers may use Remote Procedure Call (RPC) semantics for the forwarding of the client requests, and prior to the forwarding of a client request, a new unique transaction ID can substituted for an original transaction ID in the client request so that forwarded requests have unique transaction IDs.

Abstract translation: 对于每个高级协议，为服务器计算机集群设置了传输控制协议（TCP）连接的相应网格，用于转发客户端请求。 每个网格在集群中的每对服务器计算机之间具有相对的相对的TCP连接对。 例如，高级协议包括网络文件系统（NFS）协议和通用因特网文件系统（CIFS）协议。 每个网格可以在多个客户端之间共享，因为不需要为每个客户端维护单独的TCP连接状态。 服务器计算机可以使用远程过程调用（RPC）语义来转发客户端请求，并且在转发客户端请求之前，新的唯一事务ID可以替换客户端请求中的原始事务ID，从而转发请求 具有唯一的事务ID。

公开(公告)号：US20060064554A1

公开(公告)日：2006-03-23

申请号：US10945654

申请日：2004-09-21

Applicant: Stephen Fridella ,  Xiaoye Jiang ,  Uday Gupta ,  Sorin Faibish

Inventor： Stephen Fridella ,  Xiaoye Jiang ,  Uday Gupta ,  Sorin Faibish

IPC: G06F13/00

CPC classification number: G06F3/0659 ,  G06F3/061 ,  G06F3/0614 ,  G06F3/0622 ,  G06F3/0637 ,  G06F3/0683 ,  G06F9/526 ,  G06F2209/523

Abstract: A protocol is provided for allocating file locking tasks between primary and secondary data mover computers in a network file server. When there is frequent read access and infrequent write access to a file, a primary data mover grants read locks to the entire file to secondary data movers, and the secondary data movers grant read locks to clients requesting read access. When write access to the file is needed, the read locks to the entire file are released and the read locks granted to the clients are released or expire or are demoted to non-conflicting byte range locks managed by the primary data mover. Concurrent read and write access to the same file is then managed by the primary data mover.

Abstract translation: 提供了一种协议，用于在网络文件服务器中的主要和次要数据移动计算机之间分配文件锁定任务。 当对文件进行频繁的读取访问和不频繁的写入访问时，主数据移动器将授予对整个文件的读取锁定到辅助数据移动器，次要数据移动器将向读取访问权限的客户端授予读取锁定。 当需要对文件进行写入访问时，释放对整个文件的读取锁定，授予客户端的读取锁定将被释放或过期或降级到由主数据移动器管理的非冲突字节范围锁定。 然后由主数据移动器管理对同一文件的并发读写访问。

公开(公告)号：US09165014B1

公开(公告)日：2015-10-20

申请号：US13536358

申请日：2012-06-28

Applicant: Sorin Faibish ,  John M. Bent ,  Percy Tzelnic ,  Gary Grider ,  Aaron Torres

Inventor： Sorin Faibish ,  John M. Bent ,  Percy Tzelnic ,  Gary Grider ,  Aaron Torres

IPC: G06F17/30

CPC classification number: G06F17/30215 ,  G06F17/30244

Abstract: Techniques are provided for storing files in a parallel computing system using different resolutions. A method is provided for storing at least one file generated by a distributed application in a parallel computing system. The file comprises one or more of a complete file and a sub-file. The method comprises the steps of obtaining semantic information related to the file; generating a plurality of replicas of the file with different resolutions based on the semantic information; and storing the file and the plurality of replicas of the file in one or more storage nodes of the parallel computing system. The different resolutions comprise, for example, a variable number of bits and/or a different sub-set of data elements from the file. A plurality of the sub-files can be merged to reproduce the file.

Abstract translation: 提供了使用不同分辨率在并行计算系统中存储文件的技术。 提供了一种用于将由分布式应用生成的至少一个文件存储在并行计算系统中的方法。 该文件包括完整文件和子文件中的一个或多个。 该方法包括获取与文件相关的语义信息的步骤; 基于语义信息生成具有不同分辨率的文件的多个副本; 以及将所述文件的文件和所述多个副本存储在所述并行计算系统的一个或多个存储节点中。 不同的分辨率例如包括来自文件的可变数量的位和/或不同的数据元素子集。 可以合并多个子文件以再现该文件。

公开(公告)号：US08095577B1

公开(公告)日：2012-01-10

申请号：US12917910

申请日：2010-11-02

Applicant: Sorin Faibish ,  Peter Bixby ,  Uday K. Gupta ,  Walter A. O'Brien, III ,  Stephen A. Fridella

Inventor： Sorin Faibish ,  Peter Bixby ,  Uday K. Gupta ,  Walter A. O'Brien, III ,  Stephen A. Fridella

IPC: G06F17/30

CPC classification number: G06F17/30091 ,  G06F3/0685 ,  G06F3/0689

Abstract: A method is used in managing metadata. Data of an object is stored in a magnetic hard disk set in an object addressable data storage system. Metadata for the object is stored in an SSD set in the object addressable data storage system. The metadata includes information necessary for determining the location of the data of the object in the magnetic hard disk set.

Abstract translation: 一种方法用于管理元数据。 对象的数据被存储在对象可寻址数据存储系统中的磁性硬盘集中。 对象的元数据存储在目标可寻址数据存储系统中的SSD集中。 元数据包括确定磁性硬盘组中对象的数据位置所需的信息。

公开(公告)号：US07953819B2

公开(公告)日：2011-05-31

申请号：US10646851

申请日：2003-08-22

Applicant: Rui Liang ,  Jiannan Zheng ,  Mark K Ku ,  Xiaoye Jiang ,  Sorin Faibish

Inventor： Rui Liang ,  Jiannan Zheng ,  Mark K Ku ,  Xiaoye Jiang ,  Sorin Faibish

IPC: G06F15/16 ,  G06F7/00 ,  G06F17/00

CPC classification number: H04L67/1095 ,  G06F17/30067 ,  H04L29/06 ,  H04L67/1097 ,  H04L69/329

Abstract: A storage object such as a virtual disk drive or a raw logical volume is contained in a UNIX compatible file so that the file containing the storage object can be exported using the NFS or CIFS protocol and shared among UNIX and MS Windows clients or servers. The storage object can be replicated and backed up using conventional file replication and backup facilities without disruption of client access to the storage object. For client access to data of the storage object, a software driver accesses the file containing the storage object. For example, a software driver called a virtual SCSI termination is used to access a file containing a virtual SCSI disk drive. Standard storage services use the SCSI over IP protocol to access the virtual SCSI termination. An IP replication or snapshot copy facility may access the file containing the virtual SCSI disk drive using a higher-level protocol.

Abstract translation: 存储对象（如虚拟磁盘驱动器或原始逻辑卷）包含在UNIX兼容文件中，以便可以使用NFS或CIFS协议导出包含存储对象的文件，并在UNIX和MS Windows客户端或服务器之间共享。 可以使用传统的文件复制和备份功能来复制和备份存储对象，而不会中断对存储对象的客户端访问。 为了客户机访问存储对象的数据，软件驱动程序访问包含存储对象的文件。 例如，称为虚拟SCSI终端的软件驱动程序用于访问包含虚拟SCSI磁盘驱动器的文件。 标准存储服务使用SCSI over IP协议来访问虚拟SCSI终端。 IP复制或快照复制功能可以使用更高级别的协议访问包含虚拟SCSI磁盘驱动器的文件。