Reliable datagram transport service
    1.
    发明授权
    Reliable datagram transport service 失效
    可靠的数据报传输服务

    公开(公告)号:US07171484B1

    公开(公告)日:2007-01-30

    申请号:US09980759

    申请日:2000-05-24

    IPC分类号: G06F15/16

    摘要: A distributed computer system includes a source endnode including a source process which produces message data and a send work queue having work queue elements that describe the message data for sending. A destination endnode includes a destination process and a receive work queue having work queue elements that describe where to place incoming message data. A communication fabric provides communication between the source endnode and the destination endnode. An end-to-end context is provided at the source endnode and the destination endnode storing state information to ensure the reception and sequencing of message data sent from the source endnode to the destination endnode permitting reliable datagram service between the source endnode and the destination endnode.

    摘要翻译: 分布式计算机系统包括源端节点,其包括产生消息数据的源过程和具有描述用于发送的消息数据的工作队列元素的发送工作队列。 目的地节点包括目的地进程和具有描述入站消息数据的位置的工作队列元素的接收工作队列。 通信结构提供源端节点和目的端节点之间的通信。 在源端节点和目的地端节点提供端到端上下文存储状态信息,以确保从源端节点发送到目的端点的消息数据的接收和排序,从而允许源端节点与目的端节点之间的可靠数据报服务 。

    Providing a Memory Region or Memory Window Access Notification on a System Area Network
    2.
    发明申请
    Providing a Memory Region or Memory Window Access Notification on a System Area Network 有权
    在系统区域网络上提供内存区域或内存窗口访问通知

    公开(公告)号:US20120265840A1

    公开(公告)日:2012-10-18

    申请号:US13535560

    申请日:2012-06-28

    IPC分类号: G06F15/167

    CPC分类号: H04L67/1097 G06F12/1466

    摘要: A system and method for providing a memory region/memory window (MR/MW) access notification on a system area network are provided. Whenever a previously allocated MR/MW is accessed, such as via a remote direct memory access (RDMA) read/write operation, a notification of the access is generated and written to a queue data structure associated with the MR/MW. In one illustrative embodiment, this queue data structure may be a MR/MW event queue (EQ) data stricture that is created and used for all consumer processes and all MR/MWs. In other illustrative embodiments, the EQ is associated with a protection domain. In yet another illustrative embodiment, an event record may be posted to an asynchronous event handler in response to the accessing of the MR/MW. In another illustrative embodiment, a previously posted queue element may be used to generate a completion queue element in response to the accessing of the MR/MW.

    摘要翻译: 提供了一种用于在系统区域网络上提供存储器区域/存储器窗口(MR / MW)访问通知的系统和方法。 无论何时访问先前分配的MR / MW,例如通过远程直接存储器访问(RDMA)读/写操作,生成访问通知并将其写入与MR / MW相关联的队列数据结构。 在一个说明性实施例中,该队列数据结构可以是创建并用于所有消费者进程和所有MR / MW的MR / MW事件队列(EQ)数据限制。 在其他说明性实施例中,EQ与保护域相关联。 在另一说明性实施例中,可以响应于MR / MW的访问而将事件记录发布到异步事件处理程序。 在另一示例性实施例中,可以使用先前发布的队列元素来响应于MR / MW的访问来生成完成队列元素。

    Providing a memory region or memory window access notification on a system area network
    3.
    发明授权
    Providing a memory region or memory window access notification on a system area network 有权
    在系统区域网络上提供内存区域或内存窗口访问通知

    公开(公告)号:US08244826B2

    公开(公告)日:2012-08-14

    申请号:US11877337

    申请日:2007-10-23

    IPC分类号: G06F15/167

    CPC分类号: H04L67/1097 G06F12/1466

    摘要: Mechanisms for providing a memory region/memory window (MR/MW) access notification on a system area network are provided. Whenever a previously allocated MR/MW is accessed, such as via a remote direct memory access (RDMA) read/write operation, a notification of the access is generated and written to a queue data structure associated with the MR/MW. In one illustrative embodiment, this queue data structure may be a MR/MW event queue (EQ) data structure that is created and used for all consumer processes and all MR/MWs. In other illustrative embodiments, the EQ is associated with a protection domain. In yet another illustrative embodiment, an event record may be posted to an asynchronous event handler in response to the accessing of the MR/MW. In another illustrative embodiment, a previously posted queue element may be used to generate a completion queue element in response to the accessing of the MR/MW.

    摘要翻译: 提供了在系统区域网络上提供存储器区域/存储器窗口(MR / MW)访问通知的机制。 无论何时访问先前分配的MR / MW,例如通过远程直接存储器访问(RDMA)读/写操作,生成访问通知并将其写入与MR / MW相关联的队列数据结构。 在一个说明性实施例中,该队列数据结构可以是创建并用于所有消费者进程和所有MR / MW的MR / MW事件队列(EQ)数据结构。 在其他说明性实施例中,EQ与保护域相关联。 在另一说明性实施例中,可以响应于MR / MW的访问而将事件记录发布到异步事件处理程序。 在另一示例性实施例中,可以使用先前发布的队列元素来响应于MR / MW的访问来生成完成队列元素。

    Providing a memory region or memory window access notification on a system area network
    4.
    发明授权
    Providing a memory region or memory window access notification on a system area network 有权
    在系统区域网络上提供内存区域或内存窗口访问通知

    公开(公告)号:US08370447B2

    公开(公告)日:2013-02-05

    申请号:US13535560

    申请日:2012-06-28

    IPC分类号: G06F15/167

    CPC分类号: H04L67/1097 G06F12/1466

    摘要: A system and method for providing a memory region/memory window (MR/MW) access notification on a system area network are provided. Whenever a previously allocated MR/MW is accessed, such as via a remote direct memory access (RDMA) read/write operation, a notification of the access is generated and written to a queue data structure associated with the MR/MW. In one illustrative embodiment, this queue data structure may be a MR/MW event queue (EQ) data structure that is created and used for all consumer processes and all MR/MWs. In other illustrative embodiments, the EQ is associated with a protection domain. In yet another illustrative embodiment, an event record may be posted to an asynchronous event handler in response to the accessing of the MR/MW. In another illustrative embodiment, a previously posted queue element may be used to generate a completion queue element in response to the accessing of the MR/MW.

    摘要翻译: 提供了一种用于在系统区域网络上提供存储器区域/存储器窗口(MR / MW)访问通知的系统和方法。 无论何时访问先前分配的MR / MW,例如通过远程直接存储器访问(RDMA)读/写操作,生成访问通知并将其写入与MR / MW相关联的队列数据结构。 在一个说明性实施例中,该队列数据结构可以是创建并用于所有消费者进程和所有MR / MW的MR / MW事件队列(EQ)数据结构。 在其他说明性实施例中,EQ与保护域相关联。 在另一说明性实施例中,可以响应于MR / MW的访问而将事件记录发布到异步事件处理程序。 在另一示例性实施例中,可以使用先前发布的队列元素来响应于MR / MW的访问来生成完成队列元素。

    Collective send operations on a system area network
    5.
    发明授权
    Collective send operations on a system area network 失效
    在系统区域网络上集体发送操作

    公开(公告)号:US07895601B2

    公开(公告)日:2011-02-22

    申请号:US11621752

    申请日:2007-01-10

    IPC分类号: G06F15/163

    摘要: Mechanisms for collective send operations on a system area network are provided. The mechanisms of the illustrative embodiments provide for the creation, modification, and removal of collective send queues (CSQs) that allow the upper level protocol (ULP) used by a consumer to send the same message to a collective set of queue pairs (QPs). In order to use the transport services of a CSQ, a consumer process posts a write work request (WR) to the CSQ. The write WR causes a write work queue element (WQE) to be generated and placed in the CSQ. A channel interface (CI) is provided that effectively copies the write WQE to all of the send queues (SQs) of the QPs in the QP set associated with the CSQ. When all the QPs complete processing of their respective write WQEs, the HCA releases all data segments referenced by the write WR.

    摘要翻译: 提供了系统区域网络上集体发送操作的机制。 说明性实施例的机制提供了允许消费者使用的上级协议(ULP)向同一组队列(QP)发送相同消息的集体发送队列(CSQ)的创建,修改和删除, 。 为了使用CSQ的传输服务,消费者进程向CSQ发送写入工作请求(WR)。 写WR使得生成写入工作队列元素(WQE)并将其放置在CSQ中。 提供了一种通道接口(CI),其有效地将写入WQE复制到与CSQ相关联的QP集合中的QP的所有发送队列(SQ)。 当所有QP完成其各自写入WQE的处理时,HCA将释放由写入WR引用的所有数据段。

    System and Method for Providing a Memory Region or Memory Window Access Notification on a System Area Network
    6.
    发明申请
    System and Method for Providing a Memory Region or Memory Window Access Notification on a System Area Network 有权
    在系统区域网络上提供存储器区域或存储器窗口访问通知的系统和方法

    公开(公告)号:US20090106771A1

    公开(公告)日:2009-04-23

    申请号:US11877337

    申请日:2007-10-23

    IPC分类号: G06F9/46

    CPC分类号: H04L67/1097 G06F12/1466

    摘要: A system and method for providing a memory region/memory window (MR/MW) access notification on a system area network are provided. Whenever a previously allocated MR/MW is accessed, such as via a remote direct memory access (RDMA) read/write operation, a notification of the access is generated and written to a queue data structure associated with the MR/MW. In one illustrative embodiment, this queue data structure may be a MR/MW event queue (EQ) data structure that is created and used for all consumer processes and all MR/MWs. In other illustrative embodiments, the EQ is associated with a protection domain. In yet another illustrative embodiment, an event record may be posted to an asynchronous event handler in response to the accessing of the MR/MW. In another illustrative embodiment, a previously posted queue element may be used to generate a completion queue element in response to the accessing of the MR/MW.

    摘要翻译: 提供了一种用于在系统区域网络上提供存储器区域/存储器窗口(MR / MW)访问通知的系统和方法。 无论何时访问先前分配的MR / MW,例如通过远程直接存储器访问(RDMA)读/写操作,生成访问通知并将其写入与MR / MW相关联的队列数据结构。 在一个说明性实施例中,该队列数据结构可以是创建并用于所有消费者进程和所有MR / MW的MR / MW事件队列(EQ)数据结构。 在其他说明性实施例中,EQ与保护域相关联。 在另一说明性实施例中,可以响应于MR / MW的访问而将事件记录发布到异步事件处理程序。 在另一示例性实施例中,可以使用先前发布的队列元素来响应于MR / MW的访问来生成完成队列元素。

    System and Method for Collective Send Operations on a System Area Network
    7.
    发明申请
    System and Method for Collective Send Operations on a System Area Network 失效
    系统和系统区域网上集体发送操作的方法

    公开(公告)号:US20080168471A1

    公开(公告)日:2008-07-10

    申请号:US11621752

    申请日:2007-01-10

    IPC分类号: G06F3/00

    摘要: A system and method for collective send operations on a system area network are provided. The mechanisms of the illustrative embodiments provide for the creation, modification, and removal of collective send queues (CSQs) that allow the upper level protocol (ULP) used by a consumer to send the same message to a collective set of queue pairs (QPs). In order to use the transport services of a CSQ, a consumer process posts a write work request (WR) to the CSQ. The write WR causes a write work queue element (WQE) to be generated and placed in the CSQ. A channel interface (CI) is provided that effectively copies the write WQE to all of the send queues (SQs) of the QPs in the QP set associated with the CSQ. When all the QPs complete processing of their respective write WQEs, the HCA releases all data segments referenced by the write WR.

    摘要翻译: 提供了一种用于系统区域网络上的集体发送操作的系统和方法。 说明性实施例的机制提供了允许消费者使用的上级协议(ULP)向同一组队列(QP)发送相同消息的集体发送队列(CSQ)的创建,修改和删除, 。 为了使用CSQ的传输服务,消费者进程向CSQ发送写入工作请求(WR)。 写WR使得生成写入工作队列元素(WQE)并将其放置在CSQ中。 提供了一种通道接口(CI),其有效地将写入WQE复制到与CSQ相关联的QP集合中的QP的所有发送队列(SQ)。 当所有QP完成其各自写入WQE的处理时,HCA将释放由写入WR引用的所有数据段。

    Methods and devices for coupling electromagnetic radiation using diffractive optical elements
    8.
    发明授权
    Methods and devices for coupling electromagnetic radiation using diffractive optical elements 失效
    使用衍射光学元件耦合电磁辐射的方法和装置

    公开(公告)号:US07099528B2

    公开(公告)日:2006-08-29

    申请号:US10752987

    申请日:2004-01-07

    IPC分类号: G02B6/28 G02B6/26 G02B6/42

    CPC分类号: G02B6/2848 G02B6/32 G02B6/34

    摘要: Diffractive optical elements are used in methods and devices for coupling or distributing electromagnetic radiation. The diffractive optical. elements may be adapted to split one or more streams of radiation into multiple streams of radiation or to combine multiple streams of radiation to produce single streams. A plurality of diffractive optical elements may be fabricated on substrates by lithographic or molding methods. Diffractive optical elements that split one or more streams of radiation can be combined with diffractive optical elements that combine multiple streams of radiation to provide a coupling device, for example, a star coupler. The star coupler may be used in networked optical communication systems to provide multiple coupling between a plurality of optical drivers and a plurality of optical receivers. Aspects can be applied to any type of electromagnetic radiation having a wavelength that can be used as a medium for transferring information.

    摘要翻译: 衍射光学元件用于耦合或分配电磁辐射的方法和装置。 衍射光学。 元件可以适于将一个或多个辐射流分裂成多个辐射流或者组合多个辐射流以产生单个流。 多个衍射光学元件可以通过光刻或模制方法在基底上制造。 分离一个或多个辐射流的衍射光学元件可以与组合多个辐射流的衍射光学元件组合以提供耦合器件,例如星形耦合器。 星形耦合器可以用于网络光通信系统中以在多个光驱动器和多个光接收器之间提供多个耦合。 方面可以应用于具有可用作传送信息的介质的波长的任何类型的电磁辐射。

    System for processing early arrival messages within a multinode
asynchronous data communications system
    9.
    发明授权
    System for processing early arrival messages within a multinode asynchronous data communications system 失效
    用于在多节点异步数据通信系统内处理提前到达消息的系统

    公开(公告)号:US5878226A

    公开(公告)日:1999-03-02

    申请号:US853701

    申请日:1997-05-13

    CPC分类号: G06F13/387

    摘要: A message-passing protocol for accommodating early arrival messages passed between source and destination nodes in a computer system with a plurality of asynchronous computing nodes interconnected by bidirectional asynchronous communications channels. The protocol includes transmitting the message from sender to receiver without waiting for a request for the message from the receiver; determining at the receiver if a receive buffer has been posted for the message; and if the receive buffer has not been posted for the message, then either truncating the message by storing its message header in an early arrival queue at the receiver and discarding its data or allocating a temporary receive buffer at the receiver to hold the message data. Upon the receiver being ready to post a receive buffer for an early arrival message, the receiver checks the early arrival queue for the corresponding message header, and if the message header is in the early arrival queue and the message data has been discarded, then the receiver sends a pull request to the sender to retransmit the message to the receiver.

    摘要翻译: 一种消息传递协议,用于容纳在具有通过双向异步通信信道互连的多个异步计算节点的计算机系统中的源节点与目的节点之间传递的早到信息。 该协议包括从发送方发送消息到接收方,而不等待来自接收方的消息请求; 在接收器处确定是否已经为消息发布接收缓冲器; 并且如果接收缓冲器尚未被发送给消息,则通过将消息头部存储在接收器的早期到达队列中并且丢弃其数据或者在接收器处分配临时接收缓冲器来保存消息数据来截断该消息。 在接收器准备好发送用于早期到达消息的接收缓冲器的情况下,接收机检查相应消息报头的早期到达队列,并且如果消息报头处于早期到达队列中并且消息数据已被丢弃,则 接收方向发送方发送拉取请求,以将消息重发到接收方。

    MODULAR OPTICAL BACKPLANE AND ENCLOSURE
    10.
    发明申请
    MODULAR OPTICAL BACKPLANE AND ENCLOSURE 有权
    模块式光学背板和外壳

    公开(公告)号:US20150301299A1

    公开(公告)日:2015-10-22

    申请号:US13597530

    申请日:2012-08-29

    IPC分类号: G02B6/44 G02B6/43 G02B6/38

    摘要: Embodiments are directed to decomposing an all-to-all interconnection network topology into a plurality of smaller all-to-all interconnection network elements, replicating the interconnection network elements in a modular fashion, wherein the modular interconnection network elements construct the all-to-all interconnection network topology. Embodiments are directed to an apparatus comprising a shuffle cable assembly comprising a plurality of shuffle cables, where each of the plurality of shuffle cables comprises a plurality of optical fibers and a plurality of connectors, a block configured to organize, align, and maintain a position of the plurality of connectors, and at least one handle coupled to the block and configured to actuate the plurality of connectors.

    摘要翻译: 实施例旨在将全对所有互连网络拓扑分解成多个更小的全对所有互连网络元件,以模块化方式复制互连网络元件,其中模块化互连网络元件构造全对齐互连网络元件, 所有互联网络拓扑。 实施例涉及一种包括包括多个洗牌电缆的洗牌电缆组件的装置,其中多个洗牌电缆中的每一个包括多个光纤和多个连接器,被配置为组织,对准和维持位置的块 以及至少一个手柄,其联接到所述块并且被配置为致动所述多个连接器。