公开(公告)号：US20170026300A1

公开(公告)日：2017-01-26

申请号：US15184375

申请日：2016-06-16

申请人： lntel Corporation

发明人： Todd Rimmer ,  Thomas D. Lovett ,  Albert Cheng

IPC分类号： H04L12/801 ,  H04L12/925 ,  H04L12/919 ,  G06F5/06 ,  H04L12/931 ,  H04L12/861 ,  G06F13/16 ,  H04L12/851 ,  H04L12/933

CPC分类号： H04L47/39 ,  G06F5/065 ,  G06F13/1673 ,  G06F13/4221 ,  G06F2205/067 ,  H04L47/2408 ,  H04L47/2433 ,  H04L47/245 ,  H04L47/30 ,  H04L47/722 ,  H04L47/765 ,  H04L49/103 ,  H04L49/109 ,  H04L49/358 ,  H04L49/505 ,  H04L49/70 ,  H04L49/9057

摘要： Method, apparatus, and systems for implementing flexible credit exchange within high performance fabrics. Available buffer space in a receive buffer on a receive-side of a link is managed and tracked at the transmit-side of the link using credits. Peer link interfaces coupled via a link are provided with receive buffer configuration information that specifies how the receive buffer space in each peer is partitioned and space allocated for each buffer, including a plurality of virtual lane (VL) buffers. Credits are used for tracking buffer space consumption and in credits are returned from the receive-side indicating freed buffer space. The peer link interfaces exchange credit organization information to inform the other peer of how much space each credit represents. In connection with data transfer over the link, the transmit-side de-allocates credits based on an amount of buffer space to be consumed in applicable buffers in the receive buffer. Upon space being freed in the receive buffer, the receive-side returns credit ACKnowledgements (ACKs) identifying a VL for which space has been freed.

摘要翻译： 在高性能织物中实现灵活信用交换的方法，设备和系统。 在链路的接收侧的接收缓冲区中的可用缓冲区空间在链路的发送端使用信用进行管理和跟踪。 通过链路耦合的对等链路接口提供有接收缓冲器配置信息，其指定如何对每个对等体中的接收缓冲区空间进行分区，并为每个缓冲区分配空间，包括多个虚拟通道（VL）缓冲区。 积分用于跟踪缓冲区空间消耗，从接收端返回指示释放的缓冲区空间的信用。 对等链接界面交换信用组织信息，通知对方每个信用代表多少空间。 结合链路上的数据传输，发送侧基于在接收缓冲器中的适用缓冲器中消耗的缓冲区空间量去分配信用。 在接收缓冲器中释放空间时，接收端返回识别已经释放空间的VL的信用确认（ACK）。

公开(公告)号：US20160323150A1

公开(公告)日：2016-11-03

申请号：US15207706

申请日：2016-07-12

申请人： lntel Corporation

发明人： Michael Heinz ,  Todd Rimmer ,  James Kunz ,  Mark Debbage

IPC分类号： H04L12/24 ,  H04L12/753 ,  H04L12/751

CPC分类号： H04L41/12 ,  G06F9/5066 ,  H04L41/0213 ,  H04L41/046 ,  H04L41/0893 ,  H04L45/08 ,  H04L45/48

摘要： System, method, and apparatus for improving the performance of collective operations in High Performance Computing (HPC). Compute nodes in a networked HPC environment form collective groups to perform collective operations. A spanning tree is formed including the compute nodes and switches and links used to interconnect the compute nodes, wherein the spanning tree is configured such that there is only a single route between any pair of nodes in the tree. The compute nodes implement processes for performing the collective operations, which includes exchanging messages between processes executing on other compute nodes, wherein the messages contain indicia identifying collective operations they belong to. Each switch is configured to implement message forwarding operations for its portion of the spanning tree. Each of the nodes in the spanning tree implements a ratcheted cyclical state machine that is used for synchronizing collective operations, along with status messages that are exchanged between nodes. Transaction IDs are also used to detect out-of-order and lost messages.