中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

18 records (took 0.024 seconds)

    RDMA COMPLETION AND RETRANSMIT SYSTEM AND METHOD
    1.
    发明公开
    RDMA COMPLETION AND RETRANSMIT SYSTEM AND METHOD 审中-公开
    RDMA闭幕和-NEUSENDESYSTEM和程序

    公开(公告)号：EP1692621A4

    公开(公告)日：2007-09-05

    申请号：EP04812810

    申请日：2004-12-02

    Applicant: IBM

    Inventor： BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM

    IPC: G06F15/16 ,  G06F3/00 ,  G06F9/44 ,  G06F9/46 ,  G06F13/00 ,  H04L29/06 ,  H04L29/08

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

    Abstract: A system and method for maintaining ordering in completion and retransmit operations in an RDMA environment. A system is provided for handling a completion process in an remote data memory access (RDMA) environment having a RequestOut channel (16) and a ResponseOut channel (18), comprising: a descriptor list (12, 14) for each channel, wherein each descriptor list (12, 14) includes a message descriptor for each message in the channel; an update mechanism (25) for updating a message length field in the message descriptor with a sequence number of a last byte in the message whenever a channel swap occurs between the RequestOut channel (16) and the ResponseOut channel (18); an acknowledgement (Ack) completion system that examines values in a completion context (22) and compares a sequence number of a next to complete message with a last acknowledged sequence number to determine if the message should be completed; and a read request completion system that performs completion of a read request.

    Inspection of data transfer error
    5.
    发明专利
    Inspection of data transfer error 有权
    检查数据传输错误

    公开(公告)号：JP2011176835A

    公开(公告)日：2011-09-08

    申请号：JP2011053884

    申请日：2011-03-11

    Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

    Inventor： BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM ,  SHALEV LEAH

    IPC: H04L29/08 ,  H04L1/00 ,  H04L1/16

    CPC classification number: H04L1/1835 ,  H04L67/1097 ,  H04L69/16 ,  H04L69/161 ,  H04L69/163 ,  H04L69/22

    Abstract: PROBLEM TO BE SOLVED: To provide a method to process the placement and distribution of aligned DDP segment by a method which is different from placement and distribution of non-aligned DDP segment.
    SOLUTION: A network interface controller performs direct data placement for a memory when all segments connected specifically are aligned, i.e., carries out the cut-through without accessing a reassembly buffer, and moves the data through the reassembly buffer when all segments are non-aligned. The validation of CRC is performed for a majority of inbound DDP segments, then, the TCP acknowledgement response to confirm segment receiving can be transmitted. In the method to calculate the CRC value, each TCP segment starts with an aligned DDP segment, and the first two bytes of TCP payload is supposed to be an MPA length field of an MPA frame.
    COPYRIGHT: (C)2011,JPO&INPIT

    Abstract translation: 要解决的问题：提供一种通过不同于不对准DDP段的布置和分布的方法来处理对准的DDP段的布置和分布的方法。 解决方案：网络接口控制器在所有特定的段对齐时执行存储器的直接数据放置，即执行直通而不访问重新组装缓冲区，并且当所有段都是 不结盟。 对大多数入站DDP段执行CRC的验证，然后可以发送用于确认段接收的TCP确认响应。 在计算CRC值的方法中，每个TCP段以对齐的DDP段开始，TCP有效负载的前两个字节应为MPA帧的MPA长度字段。 版权所有（C）2011，JPO＆INPIT

    Method and system for memory address conversion and pinning
    6.
    发明专利
    Method and system for memory address conversion and pinning 有权
    用于存储器地址转换和引脚的方法和系统

    公开(公告)号：JP2008009982A

    公开(公告)日：2008-01-17

    申请号：JP2007166028

    申请日：2007-06-25

    Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

    Inventor： BEN-YEHUDA SHMUEL ,  SATRAN JULIAN ,  SHIMONY ILAN ,  MACHULSKY ZORIK ,  LEAH SHALEV ,  KRIEGER ORRAN YAAKOV ,  XENIDIS JAMES ,  GUTHRIDGE D SCOTT

    IPC: G06F12/10 ,  G06F13/28

    CPC classification number: G06F12/1081 ,  G06F12/126 ,  G06F2212/1016 ,  G06F2212/1044

    Abstract: PROBLEM TO BE SOLVED: To provide a method and system for memory address conversion and pinning.
    SOLUTION: The method includes attaching a memory address space identifier to a direct memory access (DMA) request, the DMA request is sent by a consumer and using a virtual address in a given address space. The method further includes looking up for the memory address space identifier to find a translation of the virtual address in the given address space used in the DMA request to a physical page frame. Provided that the physical page frame is found, pinning the physical page frame as long as the DMA request is in progress to prevent an unmapping operation of the virtual address in the given address space, and completing the DMA request, wherein the steps of attaching, looking up and pinning are centrally controlled by a host gateway.
    COPYRIGHT: (C)2008,JPO&INPIT

    Abstract translation: 要解决的问题：提供用于存储器地址转换和钉扎的方法和系统。 解决方案：该方法包括将存储器地址空间标识符附加到直接存储器访问（DMA）请求，DMA请求由消费者发送并且使用给定地址空间中的虚拟地址。 该方法还包括查找存储器地址空间标识符以找到在DMA请求中使用的给定地址空间中的虚拟地址到物理页面帧的转换。 只要找到物理页面帧，只要DMA请求正在进行，固定物理页面帧，以防止给定地址空间中虚拟地址的解映射操作，并完成DMA请求，其中附加， 查找和固定由主机网关集中控制。 版权所有（C）2008，JPO＆INPIT

    INCREASING TCP RE-TRANSMISSION PROCESS SPEED
    7.
    发明申请
    INCREASING TCP RE-TRANSMISSION PROCESS SPEED 审中-公开
    增加TCP重传过程速度

    公开(公告)号：WO2005060580A3

    公开(公告)日：2005-10-06

    申请号：PCT/US2004040758

    申请日：2004-12-06

    Applicant: IBM ,  MAKHERVAKS VADIM ,  MACHULSKY ZORIK ,  BIRAN GIORA ,  SHALEV LEAH

    Inventor： MAKHERVAKS VADIM ,  MACHULSKY ZORIK ,  BIRAN GIORA ,  SHALEV LEAH

    IPC: G06F15/16 ,  H04L29/06

    CPC classification number: H04L69/16 ,  H04L69/163

    Abstract: An RNIC implementation that performs direct data placement to memory where all segments of a particular connection are aligned, or moves data through reassembly buffers where all segments of a particular connection are non-aligned. The type of connection that cuts-through without accessing the reassembly buffers is referred to as a "Fast" connection because it is highly likely to be aligned, while the other type is referred to as a "Slow" connection. When a consumer establishes a connection, it specifies a connection type (S2). The connection type can change from Fast to Slow and back. The invention reduces memory bandwidth, latency, error recovery using TCP retransmit and provides for a "graceful recovery" from an empty receive queue. The implementation also may conduct CRC validation (S11, S6) for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (Ack) confirming segment reception.

    Abstract translation: 一个RNIC实现，执行直接数据放置到特定连接的所有段对齐的存储器，或通过重组缓冲区移动数据，其中特定连接的所有段都不对齐。 切入而不访问重组缓冲区的连接类型称为“快速”连接，因为它很可能会对齐，而另一种类型则称为“慢速”连接。 当消费者建立连接时，它指定连接类型（S2）。 连接类型可以从快速更改为慢速并返回。 本发明减少了使用TCP重传的存储器带宽，等待时间，错误恢复并且从空的接收队列提供“正常恢复”。 该实现还可以在发送确认段确认（Ack）确认段接收之前，对快速连接中的大多数入站DDP段进行CRC验证（S11，S6）。

    DATA TRANSFER ERROR CHECKING
    8.
    发明申请
    DATA TRANSFER ERROR CHECKING 审中-公开
    数据传输错误检查

    公开(公告)号：WO2005060594A3

    公开(公告)日：2007-03-01

    申请号：PCT/US2004040919

    申请日：2004-12-08

    Applicant: IBM ,  BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM ,  SHALEV LEAH

    Inventor： BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM ,  SHALEV LEAH

    IPC: H03M13/00

    CPC classification number: H04L1/1835 ,  H04L67/1097 ,  H04L69/16 ,  H04L69/161 ,  H04L69/163 ,  H04L69/22

    Abstract: A network interface controller implementation (S152) that performs direct data placement to memory where all segments of a particular connection are aligned, i.e. cuts-through without accessing reassembly buffers (S159) (i.e., a "Fast" connection type), or if all segments are non-aligned moves data through reassembly buffer (S159) (i.e., "Slow" connection type). The type of connection can change from Fast to Slow and back. The implementation preferably conducts CRC validation (S154) for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (S156) confirming segment reception. The method of computing a CRC value (S151) according to the present invention assumes that each TCP segment starts with an aligned DDP segment and that the first two bytes of a TCP payload is an MPA length field of an MPA frame.

    Abstract translation: 一种网络接口控制器实现（S152），其执行直接数据放置到特定连接的所有段对齐的存储器，即，切换而不访问重组缓冲器（S159）（即，“快速”连接类型），或者如果全部 段是通过重组缓冲器（S159）（即“慢”连接类型）的非对齐移动数据。 连接类型可以从快速变为慢速返回。 该实现优选地在发送TCP确认（S156）确认段接收之前对快速连接中的大多数入站DDP段进行CRC验证（S154）。 根据本发明的计算CRC值的方法（S151）假设每个TCP片段以对齐的DDP片段开始，并且TCP有效载荷的前两个字节是MPA帧的MPA长度字段。

    REDUCING NUMBER OF WRITE OPERATIONS RELATIVE TO DELIVERY OF OUT-OF-ORDER RDMA SEND MESSAGES
    9.
    发明申请
    REDUCING NUMBER OF WRITE OPERATIONS RELATIVE TO DELIVERY OF OUT-OF-ORDER RDMA SEND MESSAGES 审中-公开
    减少与交付不符合条件的RDMA发送消息相关的写操作数

    公开(公告)号：WO2005060579A2

    公开(公告)日：2005-07-07

    申请号：PCT/US2004040745

    申请日：2004-12-07

    Applicant: IBM ,  BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM

    Inventor： BIRAN GIORA ,  MACHULSKY ZORIK ,  MAKHERVAKS VADIM

    IPC: G06F15/167 ,  H04L29/06 ,  H04L29/08

    CPC classification number: G06F13/28 ,  H04L29/06 ,  H04L67/1097 ,  H04L69/329

    Abstract: An RNIC (16) implementation that performs direct data placement to memory where all segments of a particular connection are aligned, or moves data through reassembly buffers (34) where all segments of a particular connection are non-aligned. The type of connection that cuts-through without accessing the reassembly buffers (34) is referred to as a "Fast" connection because it is highly likely to be aligned, while the other type is referred to as a "Slow" connection. When a consumer establishes a connection, it specifies a connection type. The connection type can change from Fast to Slow and back. The invention reduces memory bandwidth, latency, error recovery using TCP retransmit and provides for a "graceful recovery" from an empty receive queue. The implementation also may conduct CRC validation for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (Ack) confirming segment reception.

    Abstract translation: 执行直接数据放置到特定连接的所有段对齐的存储器的RNIC（16）实现，或者通过重组缓冲器（34）移动数据，其中特定连接的所有段都不对齐。 在不访问重组缓冲器（34）的情况下，直接连接的类型被称为“快速”连接，因为它很可能被对准，而另一种类型被称为“慢”连接。 当消费者建立连接时，它指定一个连接类型。 连接类型可以从快速变慢到慢。 本发明减少了内存带宽，延迟，使用TCP重传的错误恢复，并提供了从空接收队列的“优雅恢复”。 在确认段接收之前，实现还可以在Fast连接中的大部分入站DDP段进行CRC校验。

Patent Agency Ranking