公开(公告)号：WO2019016908A1

公开(公告)日：2019-01-24

申请号：PCT/JP2017/026186

申请日：2017-07-20

Applicant: 三菱電機株式会社

Inventor： 桑原　茂樹

IPC: H04N5/915 ,  H04N5/919 ,  H04N5/92 ,  H04N7/18 ,  H04N21/2312

CPC classification number: H04N5/915 ,  H04N5/919 ,  H04N5/92 ,  H04N7/18 ,  H04N21/2312

Abstract: 従来の技術では、符号化された映像データを２つの記録媒体にフレーム毎に交互に記録する構成になっていたので、１つの記録媒体が故障すると奇数(または偶数)の映像フレームが欠落するためフレームレートが低下するという問題があった。本発明は、データを複数の記録装置へフレーム中の画素データを分散して記録する書き込み制御部と、記録装置が正常か故障かを判断する記録装置故障判断部と、記録装置故障判断部にていずれかの記録装置が故障と判断された場合、正常と判断された記録装置に記録されているデータを用いて故障と判断された記録装置に記録されているデータを補間し、再生データを生成する再生データ生成部とを備える。

公开(公告)号：WO2010124291A1

公开(公告)日：2010-10-28

申请号：PCT/US2010/032436

申请日：2010-04-26

Applicant: LEVEL 3 COMMUNICATIONS, LLC ,  CROWDER, William ,  KOLLER, Jeffrey ,  FULLAGAR, David

Inventor： CROWDER, William ,  KOLLER, Jeffrey ,  FULLAGAR, David

IPC: G06F15/173

CPC classification number: G06F15/173 ,  G06F17/30067 ,  H04L67/06 ,  H04L67/2842 ,  H04L67/2847 ,  H04N21/00 ,  H04N21/2225 ,  H04N21/2312 ,  H04N21/23439 ,  H04N21/2408

Abstract: Provided are computer-implemented methods and systems for performing media resource storage and management. The computer-implemented method and system implemented as a request manager is capable of monitoring requests for media resources in a content delivery network. For each monitored request, the request manager determines whether to generate a multifile for the requested media resource. For example, the request manager can first determine whether the media resource is eligible for multifile generation. If eligible, the request manager then determines whether the media resource has reached a popularity threshold. If the media resource has reached the popularity threshold, the request manager initiates generation of the multifile for the requested media resource. Generally, the generated multifile is stored in a storage system associated with the content delivery network.

Abstract translation: 提供用于执行媒体资源存储和管理的计算机实现的方法和系统。 实现为请求管理器的计算机实现的方法和系统能够监视内容传送网络中媒体资源的请求。 对于每个被监视的请求，请求管理器确定是否为所请求的媒体资源生成多媒体文件。 例如，请求管理器可以首先确定媒体资源是否符合多媒体生成的资格。 如果符合条件，则请求管理器然后确定媒体资源是否已经达到流行度阈值。 如果媒体资源已经达到流行度阈值，则请求管理器启动生成所请求的媒体资源的多媒体文件。 通常，所生成的多媒体文件存储在与内容传送网络相关联的存储系统中。

公开(公告)号：WO2008005165A2

公开(公告)日：2008-01-10

申请号：PCT/US2007013980

申请日：2007-06-13

Applicant: OMNEON VIDEO NETWORKS ,  WANIGASEKARA-MOHOTTI DON H ,  CRAIG DONALD M

Inventor： WANIGASEKARA-MOHOTTI DON H ,  CRAIG DONALD M

CPC classification number: H04L67/1097 ,  H04L67/28 ,  H04L67/2823 ,  H04L67/288 ,  H04N21/2182 ,  H04N21/2312 ,  H04N21/2343 ,  H04N21/6581

Abstract: A method for transcoding a file on a distributed file system is described. The distributed file system stores portions of the file across a plurality of distinct physical storage locations. A request to transcode the file is received. The file is transcoded from a first format to a second format using a processing unit of at least one of the physical storage locations.

Abstract translation: 描述用于对分布式文件系统上的文件进行代码转换的方法。 分布式文件系统存储多个不同物理存储位置的文件的部分。 接收到转码文件的请求。 使用物理存储位置中的至少一个的处理单元将文件从第一格式转码为第二格式。

公开(公告)号：WO2007105078A1

公开(公告)日：2007-09-20

申请号：PCT/IB2007/000612

申请日：2007-03-06

Applicant: DIBCOM ,  FALLON, Henri ,  DE MARCHI, Stéphane

Inventor： FALLON, Henri ,  DE MARCHI, Stéphane

IPC: H04N5/00

CPC classification number: H04N21/434 ,  H04N21/23106 ,  H04N21/2312 ,  H04N21/2408 ,  H04N21/84

Abstract: This method is for processing data frames (Fi) received over a communication channel, each data frame (Fi) comprising a data section (Si) for forming part of a data table and location information (Mi) associated with the data section and designating a location of said section within the data table. The data table and a metadata table (MT) are built and stored as the data frames are received, to be made available to a host processor (2). The method comprises the following steps for each received data frame: - buffering (10, 12) the data section of the received frame and the associated location information; - determining (14) an address for the data section in a table memory (8) based on the location information; - writing the data section at the determined address into the table memory; and - writing an entry of a metadata table, wherein said entry comprises the location information.

Abstract translation: 该方法用于处理通过通信信道接收的数据帧（Fi），每个数据帧（Fi）包括用于形成数据表的一部分的数据部分（Si）和与数据部分相关联的位置信息（Mi），并指定 所述部分在数据表中的位置。 数据表和元数据表（MT）在数据帧被接收时被建立和存储，以供主机处理器（2）使用。 该方法包括以下步骤： - 对接收到的帧的数据段和相关联的位置信息进行缓冲（10,12）; - 基于所述位置信息确定（14）表存储器（8）中的数据部分的地址; - 将确定的地址的数据段写入表存储器; 以及 - 写入元数据表的条目，其中所述条目包括位置信息。

公开(公告)号：WO2006002314A2

公开(公告)日：2006-01-05

申请号：PCT/US2005/022259

申请日：2005-06-22

Applicant: UNIVERSITY OF SOUTHERN CALIFORNIA ,  ZIMMERMANN, Roger ,  DESAI, Dwipal, A. ,  FU, Kun

Inventor： ZIMMERMANN, Roger ,  DESAI, Dwipal, A. ,  FU, Kun

IPC: H04N7/173

CPC classification number: H04N21/2312 ,  H04N21/2187 ,  H04N21/23103 ,  H04N21/23116 ,  H04N21/2396

Abstract: A data stream recorder system (11) for mufti-stream recording and retrieval utilizes a number of gateways (23, 24, 25), each for sending and receiving packets containing streaming multimedia content data at real-time rates via a packet data network. A session manager communicates via the network with source client devices and receiver client devices to establish and control recording and retrieval sessions. The manager assigns sessions to the gateways for the sending and receiving of the packets to and from client devices. Content is distributed across storage devices associated in storage nodes (27, 28, 29). Each of the gateways (23, 24, 25) receives packets containing content data at real time rates during a recording session and distributes the received packets from the session across all of the storage nodes (27, 28, 29). A scheduler of each respective storage node distributes content data from packets distributed to the respective storage node, across all of the digital storage devices of the respective storage node.

Abstract translation: 用于多流记录和检索的数据流记录系统（11）利用多个网关（23,24,25），每个网关用于经由分组数据网络以实时速率发送和接收包含流式多媒体内容数据的分组。 会话管理器通过网络与源客户端设备和接收者客户端设备进行通信，以建立和控制记录和检索会话。 管理员将会话分配给网关，用于向客户端设备发送和接收数据包。 内容分布在与存储节点（27,28,29）相关联的存储设备上。 每个网关（23,24,25）在记录会话期间以实时速率接收包含内容数据的分组，并且从所述会话跨所有存储节点（27,28,29）分发所接收的分组。 每个相应存储节点的调度器在分配给相应存储节点的分组中分布各自存储节点的所有数字存储设备的内容数据。

公开(公告)号：WO2004047364A3

公开(公告)日：2004-08-12

申请号：PCT/DE0303576

申请日：2003-10-28

Applicant: INFINEON TECHNOLOGIES AG ,  BENEDIX ALEXANDER ,  DANKOWSKI STEFAN ,  DUEREGGER REINHARD ,  RUF WOLFGANG

Inventor： BENEDIX ALEXANDER ,  DANKOWSKI STEFAN ,  DUEREGGER REINHARD ,  RUF WOLFGANG

IPC: G11C7/10 ,  H04N5/00 ,  H04N7/173

CPC classification number: H04N21/2182 ,  G11C7/1018 ,  G11C7/103 ,  H04N21/23116 ,  H04N21/2312 ,  H04N21/2323

Abstract: The invention relates to a memory system, in particular for network broadcasting applications such as video/audio applications. Said system comprises at least one memory (PM, PM1 to PMi), which is subdivided into several addressable memory units (M1 to Mx), each with its own output (A1 to Ax) for exchanging data. Each input (I1 to Ix) of a matrix switch is connected to a respective output (A1 to Ax) of a different memory unit. The matrix switch is operated in such a way that several of the memory units (M1 to Mx) are connected in sequential order to its outputs (OP1 to OPy), whereby a first sequence of memory units and a second sequence of memory units are connected independently to its outputs. The invention thus provides a memory system, which can service a number of requests to the same server in a deferred manner, whereby the interaction of the individual memory units (M1 to Mx) and the matrix switch (MS) enables a higher data throughput and a short access period.

Abstract translation: 一种存储系统，特别是对网络的广播应用，例如视频/音频应用中，具有至少一个存储器（PM，PM1至PMI）被划分成多个可寻址存储器单元（M1至MX），其中每个都有自己的输出 （A1至AXE）用于交换数据有。 的输入（I1至ix）的矩阵开关被连接到被连接在不同的存储器单元的各自的输出（A1至斧）。 矩阵开关被操作，使得多个与它的输出（OP1到OPY）按顺序存储单元（M1至MX）的连接，其中的存储器单元中，存储器单元的第二序列的第一序列被独立地连接到其输出端 ， 这提供了一种存储器系统，该系统可以处理数与矩阵开关（MS）请求对同一存储器的时间延迟，其中高数据吞吐量和低存取时间由所述个别存储器单元（M1至MX）的相互作用成为可能的。

公开(公告)号：WO2002087236A1

公开(公告)日：2002-10-31

申请号：PCT/US2002/012509

申请日：2002-04-19

Applicant: CONCURRENT COMPUTER CORPORATION

Inventor： ALLEGREZZA, Fred

IPC: H04N7/10

CPC classification number: G06F3/0635 ,  G06F3/0611 ,  G06F3/0689 ,  G06F13/1657 ,  H04N7/17336 ,  H04N21/2182 ,  H04N21/23103 ,  H04N21/2312 ,  H04N21/2315 ,  H04N21/2318 ,  H04N21/2323 ,  H04N21/2326 ,  H04N21/2393 ,  H04N21/2405

Abstract: Requests are received for storing/retrieving and storing data from/to a plurality of storage devices (100). A processor (300) is designed for handling each request, based, e.g., upon the load of each processor. A request for retrieving data is forwarded directly from the designated processor to the storage device via a switch, (250). Responses from the storage devices are routed directly to the designated processor via the switch (250). The switch (250) independently routes the request for retrieving data and the responses between the storage devices (100) and the processor, based on information obtained by the processor. Data provided by a designated processor is stored on the storage devices (100) via a switch (250). The switch (250) independently routes the data to be stored directly from the designated processor to the storage devices (100), based on information created by the processor. Requests and responses are exchanged between the switch (250) and the storage devices (100) via at least one high-speed network channel.

Abstract translation: 接收用于从多个存储设备（100）存储/检索和存储数据的请求。 处理器（300）被设计为基于例如每个处理器的负载来处理每个请求。 检索数据的请求直接从指定的处理器通过交换机转发到存储设备（250）。 存储设备的响应通过交换机（250）直接路由到指定的处理器。 交换机（250）基于处理器获得的信息独立地路由用于检索数据的请求和存储设备（100）和处理器之间的响应。 由指定处理器提供的数据经由开关（250）存储在存储设备（100）上。 基于由处理器创建的信息，交换机（250）将直接从指定处理器存储的数据独立地路由到存储设备（100）。 通过至少一个高速网络信道在交换机（250）和存储设备（100）之间交换请求和响应。

公开(公告)号：WO2002073391A1

公开(公告)日：2002-09-19

申请号：PCT/US2002/007658

申请日：2002-03-12

Applicant: EDGESTREAM, INC. ,  CHUNG, Randall, M.

Inventor： CHUNG, Randall, M.

IPC: G06F3/14

CPC classification number: H04N21/2312 ,  H04N21/23106

Abstract: A system and method is described for stream caching content that is being provided to one or more users over a network. The streaming content may be split into a plurality of file sections (F1, F30), which may then be further split into a number of subfiles (S1, S4). These subfiles may then be streamed to a plurality of users over a network. In one embodiment, file sections may be cached on a common caching server and accessed by subsequent users without having to wait until the entire requested file has been streamed to a preceding user.

Abstract translation: 描述了用于流缓存通过网络提供给一个或多个用户的内容的系统和方法。 流式传输内容可以被分割成多个文件部分（F1，F30），然后可以进一步将其分割成多个子文件（S1，S4）。 然后可以通过网络将这些子文件流传输到多个用户。 在一个实施例中，文件部分可以被缓存在公共高速缓存服务器上并被后续用户访问，而不必等到整个所请求的文件已经被流传输到前一个用户。

公开(公告)号：WO00004453A1

公开(公告)日：2000-01-27

申请号：PCT/US1999/015699

申请日：1999-07-13

IPC: G06F13/10 ,  G06F13/00 ,  G06F15/16 ,  G06F17/30 ,  H04L29/06 ,  H04N21/2312

CPC classification number: H04N21/2312

Abstract: Multi-media servers (10) provide clients (12) with streaming data requiring soft real-time guarantee and static data requiring a large amount of storage space. The servers (10) use a pull-mode protocol to communicate with clients through a real-time network. Separate data and control channels enhance the soft real-time capability of the server. The data channel (23) conforms to open standard protocol. A switched data link layer for the control channel (25) permits separate intrahost control messages that may be multicast and broadcast. The distributed file system (26) selects a specific data block size based upon the compression technique employed to enhance soft real-time guarantee. A hierarchical data structure combined with merging empty data blocks minimizes disk fragmentation. Data blocks are striped across multiple disks to improve disk utilization. A local buffer and a queue for both read and write requests provide support for simultaneous read and write data streams.

Abstract translation: 多媒体服务器（10）为客户端（12）提供需要软实时保证的流数据和需要大量存储空间的静态数据。 服务器（10）使用拉模式协议通过实时网络与客户端进行通信。 独立的数据和控制通道增强了服务器的软实时性能。 数据通道（23）符合开放标准协议。 用于控制信道（25）的交换数据链路层允许可以是多播和广播的单独的主机内控制消息。 分布式文件系统（26）基于用于增强软实时保证的压缩技术来选择特定的数据块大小。 结合合并空数据块的分层数据结构最大限度地减少了磁盘碎片。 数据块在多个磁盘之间进行条带化，以提高磁盘利用率。 本地缓冲区和读取和写入请求的队列都支持同时读取和写入数据流。

公开(公告)号：WO2018132382A1

公开(公告)日：2018-07-19

申请号：PCT/US2018/012973

申请日：2018-01-09

Applicant: QUALCOMM INCORPORATED

Inventor： BALASUBRAMANIAN, Srinivasan ,  ZACHARIAS, Leena ,  GHOLMIEH, Aziz ,  YANG, Yue

IPC: H04L1/18 ,  H04L1/00

CPC classification number: G06F11/1076 ,  G06F11/1443 ,  G06F11/1662 ,  G06F11/2094 ,  G06F11/2097 ,  G06F16/184 ,  G06F16/70 ,  G06F2211/1028 ,  H04L1/0003 ,  H04L1/0009 ,  H04L1/1825 ,  H04L1/1877 ,  H04L1/1887 ,  H04L65/4084 ,  H04L67/1095 ,  H04L67/1097 ,  H04L69/40 ,  H04N7/17336 ,  H04N21/2182 ,  H04N21/23103 ,  H04N21/23116 ,  H04N21/2312 ,  H04N21/2315 ,  H04N21/232 ,  H04N21/2393 ,  H04N21/2396 ,  H04N21/2405 ,  H04N21/241 ,  H04N21/262 ,  H04N21/44004 ,  H04N21/442 ,  H04N21/47202 ,  H04N21/6581 ,  H04N21/8456

Abstract: Techniques are described herein to dynamically control a number of hybrid automatic repeat requests (HARQ) transmissions based at least in part on flow characteristics of a data flow. Data included in a transport block may be grouped into one or more data flows based on a variety of factors. A set of performance benchmarks may be associated with each data flow. Flow characteristics for each data flow may be measured. A network entity may determine a number of HARQ transmissions to be transmitted during a HARQ procedure based on measured flow characteristics satisfying the performance benchmarks. For example, if a performance benchmark is not satisfied by its associated flow characteristic, the network entity may request additional HARQ transmissions during the HARQ procedure.