公开(公告)号：WO2013085501A1

公开(公告)日：2013-06-13

申请号：PCT/US2011/063662

申请日：2011-12-07

Applicant: INTEL CORPORATION ,  SAFRANEK, Robert, J. ,  BLANKENSHIP, Robert, G. ,  DAS SHARMA, Debendra

Inventor： SAFRANEK, Robert, J. ,  BLANKENSHIP, Robert, G. ,  DAS SHARMA, Debendra

IPC: G06F13/38 ,  G06F13/14

CPC classification number: G06F13/4282 ,  G06F12/0831 ,  G06F13/00 ,  G06F13/14 ,  G06F13/38 ,  G06F13/4234 ,  G06F13/426 ,  G06F15/781 ,  G06F2212/621 ,  G06F2213/0008 ,  G06F2213/0024 ,  G06F2213/0026 ,  H04L12/00

Abstract: Methods, apparatus, and systems, for transporting data units comprising multiple pieces of transaction data over high-speed interconnects. A flow control unit, called a KTI (Keizer Technology Interface) Flit, is implemented in a coherent multi-layer protocol supporting coherent memory transactions. The KTI Flit has a basic format that supports use of configurable fields to implement KTI Flits with specific formats that may be used for corresponding transactions. In one aspect, the KTI Flit may be formatted as multiple slots used to support transfer of multiple respective pieces of transaction data in a single Flit. The KTI Flit can also be configured to support various types of transactions and multiple KTI Flits may be combined into packets to support transfer of data such as cache line transfers.

Abstract translation: 用于通过高速互连传输包括多条交易数据的数据单元的方法，装置和系统。 称为KTI（Keizer Technology Interface）的流量控制单元以支持相干存储器事务的相干多层协议实现。 KTI Flit具有一种基本格式，支持使用可配置字段来实现可用于相应事务的特定格式的KTI Flits。 在一个方面，KTI Flit可以被格式化为用于支持在单个Flit中传送多个相应交易数据的多个时隙。 KTI Flit也可以配置为支持各种类型的事务，并且可以将多个KTI Flits组合成数据包，以支持数据传输，如缓存线传输。

公开(公告)号：WO2016185339A1

公开(公告)日：2016-11-24

申请号：PCT/IB2016/052765

申请日：2016-05-13

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  IBM (CHINA) INVESTMENT COMPANY LIMITED

Inventor： HATHORN, Roger ,  HOLLEY, Bret, Wayne ,  KALOS, Matthew, Joseph

IPC: G06F3/00

CPC classification number: G06F13/4027 ,  G06F3/0613 ,  G06F3/0656 ,  G06F3/067 ,  G06F5/12 ,  G06F13/1642 ,  G06F13/1673 ,  G06F13/36 ,  G06F2003/0691 ,  G06F2205/126 ,  G06F2213/0008 ,  H04L47/39

Abstract: A channel of a host computational device sends a command to transfer data to a control unit included in a storage controller. The channel of the host computational device receives a number of buffer credits from the control unit for communication with the control unit, where the number of buffer credits that is received is based on the control unit monitoring a number of transfer ready operations and a number of retry operations during a monitoring period while communicating with a plurality of channels that includes the channel.

Abstract translation: 主机计算装置的信道发送向存储控制器中包含的控制单元传送数据的命令。 主计算装置的信道从控制单元接收与控制单元进行通信的多个缓冲器信用，其中所接收的缓冲器信用的数目是基于控制单元监视多个传送就绪操作和多个 在与包括频道的多个频道通信的同时，在监视期间重试操作。

公开(公告)号：WO01075621A1

公开(公告)日：2001-10-11

申请号：PCT/US2001/009125

申请日：2001-03-21

IPC: G06F13/28 ,  H04L12/56 ,  H04L29/08 ,  H04L29/06

CPC classification number: G06F13/28 ,  G06F2213/0008 ,  H04L49/90 ,  H04L69/324

Abstract: The invention blends Fibre Channel ("FC") hardware with networking software to produce a network that allows network data to be transferred via direct memory access ("DMA") between two application buffers in computers separated by a network. During boot up, the FC network interface card ("NIC") drivers specify MTUs greater or equal to the segment size to the operating system so that data are not segmented into smaller datagrams during a network data write. During the network write, a first FC NIC sets up the send end of the DMA and sends the network headers of the data to a second FC NIC. The second FC NIC passes the network headers up through the protocol stack. The protocol stack locates and passes the application buffer address to the second FC NIC. The second FC NIC sets up the receive end of the DMA and sends a signal to the first FC NIC to start a buffer-to-buffer DMA transfer of the data. At the end of the buffer-to-buffer DMA transfer, the first FC NIC sends a signal to the second FC NIC indicating the status of the transfer. The first and second FC NICs may treat the entire data transfer as a Small Computer System Interface ("SCSI") disk transaction and use existing SCSI Assist Hardware to reduce the involvement of the host software.

Abstract translation: 本发明将光纤通道（“FC”）硬件与网络软件相结合，以产生网络，允许网络数据通过由网络分离的计算机中的两个应用程序缓冲区之间的直接存储器访问（“DMA”）进行传输。 在启动过程中，FC网络接口卡（“NIC”）驱动程序为操作系统指定大于或等于段大小的MTU，以便在网络数据写入期间数据不会分段成较小的数据报。 在网络写入期间，第一个FC NIC设置DMA的发送端，并将数据的网络头部发送到第二个FC NIC。 第二个FC NIC通过协议栈传递网络头。 协议栈定位并将应用程序缓冲区地址传递给第二个FC NIC。 第二个FC NIC设置DMA的接收端，并向第一个FC NIC发送一个信号，以启动数据的缓冲区到缓冲区DMA传输。 在缓冲区到缓冲区DMA传输结束时，第一个FC NIC向第二个FC NIC发送一个信号，指示传输的状态。 第一个和第二个FC NIC可以将整个数据传输视为小型计算机系统接口（“SCSI”）磁盘事务，并使用现有的SCSI Assist硬件来减少主机软件的参与。