公开(公告)号：US11063884B2

公开(公告)日：2021-07-13

申请号：US16554064

申请日：2019-08-28

Applicant: Intel Corporation

Inventor： Ilango Ganga ,  Alain Gravel ,  Thomas Lovett ,  Radia Perlman ,  Greg Regnier ,  Anil Vasudevan ,  Hugh Wilkinson

IPC: H04L12/939 ,  H04L1/16 ,  H04L12/947 ,  H04L12/861

Abstract: This disclosure describes enhancements to Ethernet for use in higher performance applications like Storage, HPC, and Ethernet based fabric interconnects. This disclosure provides various mechanisms for lossless fabric enhancements with error-detection and retransmissions to improve link reliability, frame pre-emption to allow higher priority traffic over lower priority traffic, virtual channel support for deadlock avoidance by enhancing Class of service functionality defined in IEEE 802.1Q, a new header format for efficient forwarding/routing in the fabric interconnect and header CRC for reliable cut-through forwarding in the fabric interconnect. The enhancements described herein, when added to standard and/or proprietary Ethernet protocols, broadens the applicability of Ethernet to newer usage models and fabric interconnects that are currently served by alternate fabric technologies like Infiniband, Fibre Channel and/or other proprietary technologies, etc.

公开(公告)号：US10205667B2

公开(公告)日：2019-02-12

申请号：US15614455

申请日：2017-06-05

Applicant: Intel Corporation

Inventor： Ilango Ganga ,  Alain Gravel ,  Thomas D. Lovett ,  Radia Perlman ,  Greg Regnier ,  Anil Vasudevan ,  Hugh Wilkinson

IPC: H04L12/851

Abstract: One embodiment provides a method for enabling class-based credit flow control for a network node in communication with a link partner using an Ethernet communications protocol. The method includes receiving a control frame from the link partner. The control frame includes at least one field for specifying credit for at least one traffic class and the credit is based on available space in a receive buffer associated with the at least one traffic class. The method further includes sending data packets to the link partner based on the credit, the data packets associated with the at least one traffic class.

公开(公告)号：US10791058B2

公开(公告)日：2020-09-29

申请号：US15625592

申请日：2017-06-16

Applicant: Intel Corporation

Inventor： Radia Perlman

IPC: H04L12/851 ,  H04L12/801

Abstract: Systems and methods may provide for determining a local traffic quota for a service associated with an overlay network and determining an allocation of the local traffic quota across a set of data sources associated with the overlay network. Additionally, the allocation may be imposed on one or more packets received from the set of data sources. In one example, imposing the allocation on the one or more packets includes sending the one or more packets to a parent node connected to the overlay router in a hierarchy of the overlay network if delivery of the one or more packets to the parent node complies with the allocation and delaying delivery of the one or more packets to the parent node if the packets do not comply with the allocation.

公开(公告)号：US09088511B2

公开(公告)日：2015-07-21

申请号：US13681136

申请日：2012-11-19

Applicant: Intel Corporation

Inventor： Radia Perlman

IPC: H04L12/721 ,  H04L12/703

CPC classification number: H04L45/26 ,  H04L12/4633 ,  H04L45/38

Abstract: Methods, apparatus and network architectures relating to the use of a Hop-by-Hop packet forwarding technique using “stepping stone” switches. The network architectures include use of stepping stone switches interspersed with non-stepping stone switches such as conventional network switches comprising network elements such switches, routers, repeaters, etc. The stepping stone switches are configured to route packets as multiplexed flows along tunneled sub-paths between stepping stone switches in a hop-by-hop manner with error recovery, as opposed to conventional routing under which packets are routed from a source to a destination using an arbitrary path or along a (generally) lengthy flow-based path. Accordingly, packets from a source endpoint are routed to a destination endpoint via multiple sub-paths connecting pairs of stepping stone switches, with each sub-path traversing one or more conventional switches and constituting a logical Hop in the Hop-by-Hop route.

Abstract translation: 与使用“踏脚石”开关的Hop-by-Hop分组转发技术相关的方法，装置和网络架构。 网络架构包括使用散布有非步进石开关的踏脚石开关，例如包括诸如开关，路由器，中继器等网元的常规网络交换机。踏脚石开关被配置为沿着隧道化子路径将分组路由为多路复用流 在传统的路由下，使用任意路径或沿着（通常为漫长的基于流的路径）将数据包从源路由到目的地而不是以错误恢复的逐跳方式进行。 因此，来自源端点的分组经由连接成对的踏脚石交换机的多个子路径被路由到目的地端点，每个子路径遍历一个或多个常规交换机，并且在Hop-by-Hop路由中构成逻辑Hop。

公开(公告)号：US08989017B2

公开(公告)日：2015-03-24

申请号：US13715669

申请日：2012-12-14

Applicant: Intel Corporation

Inventor： Ygdal Naouri ,  Radia Perlman

IPC: H04L29/06 ,  H04L12/26 ,  H04L12/803 ,  H04L12/801 ,  H04L12/931

CPC classification number: H04L47/122 ,  H04L47/10 ,  H04L49/351 ,  H04L49/50

Abstract: Methods, apparatus, and networks configured to manage network congestion using packet recirculation. The networks employ network elements (e.g., Rbridges in Layer 2 networks and switches/routers in Layer 3 networks) that are configured to support multi-path forwarding under which packets addressed to the same destination may be routed via multiple paths to the destination. In response to network congestion conditions, such as lack of availability of a non-congested port via which a shortest path to the destination may be accessed, a packet may be routed backward toward a source node or forwarded toward a destination along a non-shortest path. The network elements may employ loopback buffers for looping packets back toward a source via the same link the packet is received on.

Abstract translation: 配置为使用分组再循环来管理网络拥塞的方法，装置和网络。 网络采用被配置为支持多径转发的网络元件（例如，第2层网络中的R桥和第3层网络中的交换机/路由器），在该路径转发下，寻址到相同目的地的分组可以经由多条路径被路由到目的地。 响应于网络拥塞状况，诸如可能访问到目的地的最短路径的非拥塞端口的可用性不足，分组可以朝向源节点反向路由，或者沿着非最短路径朝向目的地转发 路径。 网络元件可以采用环回缓冲器，用于经由分组被接收的相同链路将分组循环回源。

公开(公告)号：US20140207928A1

公开(公告)日：2014-07-24

申请号：US13993374

申请日：2011-12-29

Applicant: INTEL CORPORATION

Inventor： Michael P. Mesnier ,  Radia Perlman

IPC: H04L12/24

CPC classification number: H04L41/507 ,  H04L41/5019 ,  H04L43/028 ,  H04L67/1097 ,  H04L67/322

Abstract: In accordance with some embodiments, identification of transport streams facilitates the classification of those streams. Classification of those streams in turn enables a classification to be matched to a quality of service policy. Thus, quality of service policies may be enforced so that different streams can be afforded appropriate quality of service.

Abstract translation: 根据一些实施例，传输流的识别便于对这些流的分类。 这些流的分类又使分类符合服务质量政策。 因此，可以执行服务质量策略，使得不同的流可以被提供适当的服务质量。

公开(公告)号：US09917891B2

公开(公告)日：2018-03-13

申请号：US14040125

申请日：2013-09-27

Applicant: Intel Corporation

Inventor： Radia Perlman

IPC: H04L29/08

CPC classification number: H04L67/1034 ,  H04L67/1027

Abstract: Methods, apparatus, and systems for distributing network loads in a manner that is resilient to system topology changes. Distribution functions and associated operations are implemented on multiple load splitters such that if a load splitter becomes inoperative, another or other load splitters can forward packets corresponding to flows previously handled by the inoperative load splitter without requiring flow state synchronization to be maintained across load splitters. The distribution functions are implemented in a manner that distributes packets for the same flows to the same servers through system topology changes, addressing both situations when servers fail and/or are taken off-line and when such servers or replacement servers are brought back on-line. The techniques are facilitated, in part, via use of redistributed flow lists and/or Bloom filters that are marked to track redistributed flows. A novel Bloom filter recycle scheme is also disclosed.

公开(公告)号：US09716660B2

公开(公告)日：2017-07-25

申请号：US14566960

申请日：2014-12-11

Applicant: Intel Corporation

Inventor： Radia Perlman

IPC: H04L12/851 ,  H04L12/801

CPC classification number: H04L47/2425 ,  H04L47/12

Abstract: Systems and methods may provide for determining a local traffic quota for a service associated with an overlay network and determining an allocation of the local traffic quota across a set of data sources associated with the overlay network. Additionally, the allocation may be imposed on one or more packets received from the set of data sources. In one example, imposing the allocation on the one or more packets includes sending the one or more packets to a parent node connected to the overlay router in a hierarchy of the overlay network if delivery of the one or more packets to the parent node complies with the allocation and delaying delivery of the one or more packets to the parent node if the packets do not comply with the allocation.

公开(公告)号：US09577791B2

公开(公告)日：2017-02-21

申请号：US13705491

申请日：2012-12-05

Applicant: Intel Corporation

Inventor： Eliel Louzoun ,  Radia Perlman ,  Ben-Zion Friedman ,  Ygdal Naouri ,  Eliezer Tamir

IPC: H04L12/28 ,  H04L1/16 ,  H04L1/18 ,  H04L1/00

CPC classification number: H04L1/1657 ,  H04L1/188 ,  H04L2001/0092

Abstract: Methods and apparatus for implementing notification by network elements of packet drops. In response to determining a packet is to be dropped, a network element such as a switch or router determines the source of the packet and returns a dropped packet notification message to the source. Upon receipt of notification, networking software or embedded hardware on the source causes the dropped packet to be retransmitted. The notification may also be sent from the network element to the destination computer to inform networking software or embedded logic implemented by the destination computer that the packet was dropped and notification to the source has been sent, thus alleviating the destination from needing to send a Selective ACKnowledge (SACK) message to inform the source the packet was not delivered. (Too narrow)

Abstract translation: 用于通过分组丢弃的网元实现通知的方法和装置。 响应于确定要丢弃分组，诸如交换机或路由器的网络元件确定分组的源，并将丢弃的分组通知消息返回给源。 收到通知后，源上的网络软件或嵌入式硬件会导致丢弃的数据包被重新发送。 该通知还可以从网络元件发送到目的地计算机，以通知目的地计算机实现的网络软件或嵌入式逻辑，该分组被丢弃并且向源的通知已经被发送，从而减轻目的地需要发送选择性 确认（SACK）消息以通知源未传送数据包。 （太窄）

公开(公告)号：US09923677B2

公开(公告)日：2018-03-20

申请号：US14583337

申请日：2014-12-26

Applicant: Intel Corporation

Inventor： Radia Perlman

IPC: H04L29/06 ,  H04L12/46 ,  H04L29/08 ,  H04L5/00

CPC classification number: H04L5/0001 ,  H04L12/4633 ,  H04L67/02 ,  H04L67/28 ,  H04L67/2842 ,  H04L69/161 ,  H04L69/163

Abstract: Methods and apparatus for multiplexing many client streams over a single connection. A proxy server establishes multiple TCP connections with respective clients that desire to access a web server connected to the proxy server via a multiplexed TCP connection. TCP packets received from the clients via the TCP connections are separated out based on their TCP connections, a packet payload data is extracted and added to client data streams. Data segments comprising sequential runs of bits from the client data streams and embedded in multiplexed (MUX) TCP packets that are sent over the multiplexed TCP connection. Upon receipt, the web server de-encapsulates the data segments and buffers them in queues allocated for each TCP connection in re-assembled client data streams. This enables the packet flows transported over the multiplexed connection for the TCP connections to be individually controlled. The multiplexed TCP connection may also be used for forwarding packet payload data generated at the web server to the clients via the proxy server and the client's TCP connections.