SUPPORTING COORDINATED ACCESS TO A FILE SYSTEM'S SHARED STORAGE USING AUTOMATIC ALIGNMENT OF A PARALLEL FILE ACCESS PROTOCOL AND METADATA MANAGEMENT
    1.
    发明申请
    SUPPORTING COORDINATED ACCESS TO A FILE SYSTEM'S SHARED STORAGE USING AUTOMATIC ALIGNMENT OF A PARALLEL FILE ACCESS PROTOCOL AND METADATA MANAGEMENT 审中-公开
    使用自动对齐的并行文件访问协议和元数据管理支持对文件系统的共享存储的协调访问

    公开(公告)号:WO2014049483A2

    公开(公告)日:2014-04-03

    申请号:PCT/IB2013058529

    申请日:2013-09-13

    CPC classification number: G06F17/30079 G06F17/30008 G06F17/30197

    Abstract: Embodiments of the invention relate to supporting coordinated access to a file system's shared disk storage subsystem using an automatic alignment of a parallel file access protocol and metadata management on a per object basis. Active file servers in the cluster have equivalent access to shared data of the file system. A file server (among servers in the cluster) receives a client's request to access a file stored by the file system in the storage subsystem using a parallel file access protocol. If the file system has selected a file server to serve as a metadata manager of said file, the file server uses the selected file server to coordinate access to the file's metadata. If the file system has not selected a file server to serve as the metadata manager of the file, the file system selects the file server that received said request as the metadata manager for said file until a file close request operation is received. The file server then performs a file open request operation for the file, to the file server selected to serve as the metadata manager. The file server then opens the file and creates a file handle of the file. The file server then encodes a file server ID, of the file server selected as the metadata manager of said file, in the file handle such that the file system's file servers may identify which file server is the file's metadata manager. The file server then returns the file handle to the client for performing I/O to said file.

    Abstract translation: 本发明的实施例涉及使用基于每个对象的自动对齐并行文件访问协议和元数据管理来支持对文件系统的共享磁盘存储子系统的协调访问。 群集中的活动文件服务器可以等效访问文件系统的共享数据。 文件服务器(群集中的服务器中)使用并行文件访问协议接收客户端的请求,以访问存储子系统中文件系统存储的文件。 如果文件系统选择文件服务器作为所述文件的元数据管理器,则文件服务器使用选择的文件服务器来协调对文件元数据的访问。 如果文件系统没有选择文件服务器作为文件的元数据管理器,则文件系统选择接收到所述请求的文件服务器作为所述文件的元数据管理器,直到接收到文件关闭请求操作为止。 文件服务器然后对该文件执行文件打开请求操作,该文件服务器被选择用作元数据管理器。 文件服务器然后打开文件并创建文件的文件句柄。 然后,文件服务器将文件服务器ID编码为文件句柄中选择作为所述文件的元数据管理器的文件服务器,使得文件系统的文件服务器可以识别哪个文件服务器是文件的元数据管理器。 文件服务器然后将文件句柄返回给客户端以对所述文件执行I / O。

    DATA INTEGRITY VALIDATION IN STORAGE SYSTEMS
    2.
    发明申请
    DATA INTEGRITY VALIDATION IN STORAGE SYSTEMS 审中-公开
    存储系统中的数据一致性验证

    公开(公告)号:WO2008138768A3

    公开(公告)日:2009-06-04

    申请号:PCT/EP2008055361

    申请日:2008-04-30

    Abstract: Data validation systems and methods are provided. Data is recorded in N data chunks on one or more storage mediums. A first validation chunk independently associated with said N data chunks comprises first validation information for verifying accuracy of data recorded in said N data chunks. The first validation chunk is associated with a first validation appendix comprising second validation information, wherein the first validation appendix is stored on a first storage medium independent of said one or more storage mediums.

    Abstract translation: 提供数据验证系统和方法。 数据记录在一个或多个存储介质上的N个数据块中。 独立地与所述N个数据块相关联的第一验证块包括用于验证记录在所述N个数据块中的数据的精度的第一验证信息。 第一验证块与包括第二验证信息的第一验证附录相关联,其中第一验证附录存储在独立于所述一个或多个存储介质的第一存储介质上。

    SYSTEM AND METHOD FOR EFFICIENTLY TRANSFERRING DATA STREAM BY MULTIMEDIA SYSTEM

    公开(公告)号:JPH10224767A

    公开(公告)日:1998-08-21

    申请号:JP32100997

    申请日:1997-11-21

    Applicant: IBM

    Abstract: PROBLEM TO BE SOLVED: To improve the throughput of a data stream by accessing a general file system interface in order to execute the zero copy transfer of a data stream from a producer module to a consumer module. SOLUTION: A video couping sub-system(VCS) 224 sets up a pipeline between the producer module 222 and a consumer module 226 and starts a data flow. The VCS 224 supplies reference to the data stored in a shared memory 212 to a consumer device 214. Thereby one copy is left in the memory 212 and data are transferred from a large capacity storage device 210 to the device 214. The contents of the device 214 are mapped in local devices such as a network and a video decoder in a server. These devices distribute digital data to consumers or distribute analog data to a cable through a high band width ATM switch or the like.

    VIDEO OPTIMIZED MEDIA STREAMER DATA FLOW ARCHITECTURE

    公开(公告)号:CA2154038A1

    公开(公告)日:1996-03-09

    申请号:CA2154038

    申请日:1995-07-17

    Applicant: IBM

    Abstract: A media streamer (10) includes at least one storage node (16, 17) comprising mass storage for storing a digital representation of at least one video presentation; and a plurality of communication nodes (14) each having at least one input port that is coupled to an output of the at least one storage node for receiving a digital representation of a video presentation therefrom. Each of the plurality of communication nodes further includes a plurality of output ports, individual ones of the plurality of output ports being operable for transmitting a digital representation as a data stream to a consumer of the digital representation. Individual ones of the output ports also have an associated data buffer for buffering a portion of a digital representation prior to a transmission of t he digital representation. Each of the plurality of communication nodes further include s a scheduler for scheduling, for individual ones of the data buffers, a read request to t he at least one storage node for a next sequential portion of the digital representation for storage within individual ones of the data buffers . Each read request for a particular ind ividual one of the data buffers is scheduled at a predetermined time such that a requested next sequential portion of the digital representation is available to a requesting communica tions node prior to a time that the requested next sequential portion is required to be trans mitted from the output port that is associated with the data buffer for which the read reque st was scheduled. The at least one storage node further includes a scheduler for sc heduling a read request to the mass storage such that a next sequential portion of a di gital representation is available for outputting, prior to the at least one storag e node receiving a read request for the next sequential portion from one of the communication nodes.

    5.
    发明专利
    未知

    公开(公告)号:DE69515153D1

    公开(公告)日:2000-03-30

    申请号:DE69515153

    申请日:1995-09-05

    Applicant: IBM

    Abstract: A video server system or media streamer 10 includes at least one control node 18; at least one storage node 16, 17 with mass storage for storing a digital representation of a video presentation; and a plurality of communication nodes 14 switchably coupled under the direction of the control node to the at least one storage node for receiving a video presentation therefrom. Each of the plurality of communication nodes is coupled to an adapter 212 by a communications bus 210, outputting a video presentation as a sequence of data bursts over the communications bus to the adapter, which converts the received sequence of data bursts to a substantially isochronous data stream. Each of the plurality of communication nodes further includes a plurality of output ports, each having an associated data buffer for buffering a portion of a digital representation prior to a transmission of the digital representation. Each of the plurality of communication nodes further includes a scheduler for scheduling, for individual data buffers, a read request to the storage nodes for a next sequential portion of the digital representation for storage within the data buffers. Each read request is scheduled at a predetermined time such that a requested next sequential portion of the digital representation is available prior to a time that the requested next sequential portion is required to be transmitted from the output port that is associated with the data buffer. The at least one storage node further includes a scheduler for scheduling a read request to the mass storage such that a next sequential portion of a digital representation is available for outputting, prior to the at least one storage node receiving a read request from one of the communication nodes.

    VIDEO OPTIMIZED MEDIA STREAMER DATA FLOW ARCHITECTURE

    公开(公告)号:CA2154038C

    公开(公告)日:2002-07-09

    申请号:CA2154038

    申请日:1995-07-17

    Applicant: IBM

    Abstract: A media streamer (10) includes at least one storage node (16, 17) comprising mass storage for storing a digital representation of at least one video presentation; and a plurality of communication nodes (14) each having at least one input port that is coupled to an output of the at least one storage node for receiving a digital representation of a video presentation therefrom. Each of the plurality of communication nodes further includes a plurality of output ports, individual ones of the plurality of output ports being operable for transmitting a digital representation as a data stream to a consumer of the digital representation. Individual ones of the output ports also have an associated data buffer for buffering a portion of a digital representation prior to a transmission of t he digital representation. Each of the plurality of communication nodes further include s a scheduler for scheduling, for individual ones of the data buffers, a read request to t he at least one storage node for a next sequential portion of the digital representation for storage within individual ones of the data buffers . Each read request for a particular ind ividual one of the data buffers is scheduled at a predetermined time such that a requested next sequential portion of the digital representation is available to a requesting communica tions node prior to a time that the requested next sequential portion is required to be trans mitted from the output port that is associated with the data buffer for which the read reque st was scheduled. The at least one storage node further includes a scheduler for sc heduling a read request to the mass storage such that a next sequential portion of a di gital representation is available for outputting, prior to the at least one storag e node receiving a read request for the next sequential portion from one of the communication nodes.

    7.
    发明专利
    未知

    公开(公告)号:DE69515153T2

    公开(公告)日:2000-09-21

    申请号:DE69515153

    申请日:1995-09-05

    Applicant: IBM

    Abstract: A video server system or media streamer 10 includes at least one control node 18; at least one storage node 16, 17 with mass storage for storing a digital representation of a video presentation; and a plurality of communication nodes 14 switchably coupled under the direction of the control node to the at least one storage node for receiving a video presentation therefrom. Each of the plurality of communication nodes is coupled to an adapter 212 by a communications bus 210, outputting a video presentation as a sequence of data bursts over the communications bus to the adapter, which converts the received sequence of data bursts to a substantially isochronous data stream. Each of the plurality of communication nodes further includes a plurality of output ports, each having an associated data buffer for buffering a portion of a digital representation prior to a transmission of the digital representation. Each of the plurality of communication nodes further includes a scheduler for scheduling, for individual data buffers, a read request to the storage nodes for a next sequential portion of the digital representation for storage within the data buffers. Each read request is scheduled at a predetermined time such that a requested next sequential portion of the digital representation is available prior to a time that the requested next sequential portion is required to be transmitted from the output port that is associated with the data buffer. The at least one storage node further includes a scheduler for scheduling a read request to the mass storage such that a next sequential portion of a digital representation is available for outputting, prior to the at least one storage node receiving a read request from one of the communication nodes.

    Unterstützen eines koordinierten Zugriffs auf einen gemeinsam genutzten Speicher eines Dateisystems unter Verwendung einer automatischen Ausrichtung eines Protokolls für einen parallelen Dateizugriff und einer Metadatenverwaltung

    公开(公告)号:DE112013004805T5

    公开(公告)日:2015-06-03

    申请号:DE112013004805

    申请日:2013-09-13

    Applicant: IBM

    Abstract: Ausführungsformen der Erfindung beziehen sich auf ein Unterstützen eines koordinierten Zugriffs auf ein gemeinsam genutztes Plattenspeicherteilsystem eines Dateisystems unter Verwendung einer automatischen Ausrichtung eines Protokolls für einen parallelen Dateizugriff und einer Metadatenverwaltung auf einer Objektgrundlage, Aktive Datei-Server in dem Cluster haben einen vergleichbaren Zugriff auf gemeinsam genutzte Daten des Dateisystems. Ein Datei-Server (aus mehreren Servern in dem Cluster) empfängt eine Client-Anforderung, unter Verwendung eines Protokolls für einen parallelen Dateizugriff auf eine Datei zuzugreifen, die durch das Dateisystem in dem Speicherteilsystem gespeichert wird. Wenn das Dateisystem einen Datei-Server ausgewählt hat, der als ein Metadaten-Manager der Datei dienen soll, koordiniert der Datei-Server anhand des ausgewählten Datei-Servers den Zugriff auf die Metadaten der Datei. Wenn das Dateisystem keinen Datei-Server für eine Funktion als Metadaten-Manager der Datei ausgewählt hat, wählt das Dateisystem den Datei-Server, der die Anforderung empfangen hat, als Metadaten-Manager für die Datei aus, bis eine Anforderungsoperation einer Dateischließung empfangen wird. Danach führt der Datei-Server für die Datei eine Operation einer Dateiöffnungsanforderung auf dem Datei-Server durch, der für eine Funktion als Metadaten-Manager ausgewählt wurde. Der Datei-Server öffnet daraufhin die Datei und erzeugt eine Dateikennung der Datei. Im Anschluss daran codiert der Datei-Server in der Dateikennung eine Datei-Server-Kennung des Datei-Servers, der als der Metadaten-Manager der Datei ausgewählt wurde, so dass die Datei-Server des Dateisystems erkennen können, welcher Datei-Server der Metadaten-Manager der Datei ist. Danach gibt der Datei-Server die Dateikennung an den Client zurück, um eine Eingabe/Ausgabe für die Datei auszuführen.

    9.
    发明专利
    未知

    公开(公告)号:DE69733305D1

    公开(公告)日:2005-06-23

    申请号:DE69733305

    申请日:1997-11-28

    Applicant: IBM

    Abstract: A general purpose interface is provided to a file system of a multimedia server for improving datastream throughput by avoiding data copying and context switching. The file system includes an export registration service providing function pointers to the file system's implementation of a predefined set of zero copy interfaces. These interfaces include an initialization service to set up necessary data structures and to place use counts on objects to prevent them from being removed while zero copy operations are in progress. A termination service in the zero copy interfaces cleans up data structures and removes use counts. Additional services validate that a specified file is suitable for zero copy operations, find and record file attributes needed to later acquire buffers at interrupt level. A data structure is defined to hold these attributes for subsequent calls. Another service locks a buffer at an interrupt level and returns the amount of data present, and provides addressability to the buffer in another address space context. A data structure holds these buffer attributes for subsequent calls. A service is additionally provided to unlock the buffer at interrupt level. Finally, a service provides information on changes to the access pattern of the file such as would be needed to support VCR functions. A hash chain emplyoing a singly linked list updated atomically allows for safe buffer lookups at interrupt level. Compare-and-swap locking faciliates this. A mailbox interface useful at interrupt leve provides prefetch and release information to the daemon. In case of a lock collision or where the buffer is not available, the buffer is acquired in a process context.

    10.
    发明专利
    未知

    公开(公告)号:AT189944T

    公开(公告)日:2000-03-15

    申请号:AT95306239

    申请日:1995-09-05

    Applicant: IBM

    Abstract: A video server system or media streamer 10 includes at least one control node 18; at least one storage node 16, 17 with mass storage for storing a digital representation of a video presentation; and a plurality of communication nodes 14 switchably coupled under the direction of the control node to the at least one storage node for receiving a video presentation therefrom. Each of the plurality of communication nodes is coupled to an adapter 212 by a communications bus 210, outputting a video presentation as a sequence of data bursts over the communications bus to the adapter, which converts the received sequence of data bursts to a substantially isochronous data stream. Each of the plurality of communication nodes further includes a plurality of output ports, each having an associated data buffer for buffering a portion of a digital representation prior to a transmission of the digital representation. Each of the plurality of communication nodes further includes a scheduler for scheduling, for individual data buffers, a read request to the storage nodes for a next sequential portion of the digital representation for storage within the data buffers. Each read request is scheduled at a predetermined time such that a requested next sequential portion of the digital representation is available prior to a time that the requested next sequential portion is required to be transmitted from the output port that is associated with the data buffer. The at least one storage node further includes a scheduler for scheduling a read request to the mass storage such that a next sequential portion of a digital representation is available for outputting, prior to the at least one storage node receiving a read request from one of the communication nodes.

Patent Agency Ranking