公开(公告)号：US20170046306A1

公开(公告)日：2017-02-16

申请号：US15337872

申请日：2016-10-28

Applicant: Huawei Technologies Co., Ltd.

Inventor： Nan LI ,  Linchun WANG ,  Hongfei CHEN

IPC: G06F15/80 ,  G06F9/38 ,  G06F15/163

CPC classification number: G06F15/80 ,  G06F9/38 ,  G06F9/3826 ,  G06F9/3828 ,  G06F13/1605 ,  G06F13/4031 ,  G06F15/163 ,  G06F15/17 ,  G06F15/825 ,  G06F2207/3868

Abstract: Disclosed are a data processing method, a processor, and a data processing device. The method comprises: an arbiter sends data D(a,1) to a first processing circuit; the first processing circuit processes the data D(a,1) to obtain data D(1,2), the first processing circuit being a processing circuit among m processing circuits; the first processing circuit sends the data D(1,2) to a second processing circuit; the second processing circuit to an mth processing circuit separately process the received data; and the arbiter receives data D(m,a) sent by the mth processing circuit. The processor further comprises an (m+1)th processing circuit. Each processing circuit in the first processing circuit to the (m+1)th processing circuit can receive first data to be processed sent by the arbiter, and process the first data to be processed. The scheme is helpful to improve efficiency of data processing.

Abstract translation: 公开了一种数据处理方法，处理器和数据处理装置。 该方法包括：仲裁器将数据D（a，1）发送到第一处理电路; 第一处理电路处理数据D（a，1）以获得数据D（1,2），第一处理电路是m个处理电路中的处理电路; 第一处理电路将数据D（1,2）发送到第二处理电路; 对第m处理电路的第二处理电路分别处理接收到的数据; 并且仲裁器接收由第m个处理电路发送的数据D（m，a）。 处理器还包括第（m + 1）个处理电路。 第（m + 1）处理电路的第一处理电路中的每个处理电路可以接收由仲裁器发送的待处理的第一数据，并处理第一待处理数据。 该方案有助于提高数据处理的效率。

公开(公告)号：US20220166859A1

公开(公告)日：2022-05-26

申请号：US17668843

申请日：2022-02-10

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yariv BRIMA ,  Yoni BICK ,  Nan LI ,  Shunfang LI

IPC: H04L69/22 ,  H04L49/101 ,  H04L49/00

Abstract: A partial packet context builder can determine a partial packet context associated with a data packet based upon packet context associated with the data packet, the partial packet context including a plurality of partial packet context fields, where a network packet processor including an action unit, the action unit including the partial packet context builder, a systolic array of arithmetic-logical units (ALUs), and a packet context builder, the packet context including a plurality of packet context fields. The systolic array of ALUs can process the partial packet context to provide a processed partial packet context, the processed partial packet context including a plurality of processed partial packet context fields. The packet context builder can merge the processed partial packet context with the packet context to provide a processed packet context.

公开(公告)号：US20220094630A1

公开(公告)日：2022-03-24

申请号：US17541437

申请日：2021-12-03

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yan ZHAO ,  Nan LI

IPC: H04L45/24 ,  H04L47/125 ,  H04L47/17

Abstract: According to a load sharing method and a network device, traffic distribution ratios are correspondingly preset for different paths. For example, an equivalent resistance method is used as a load sharing algorithm provided in embodiments of this application, to calculate a corresponding traffic distribution ratio for each possible path. The traffic distribution ratio and the corresponding path are preset for a corresponding node. In this way, traffic entering a network device is distributed to each path in the network device based on the preset ratio, without causing excessive bandwidth pressure on a shortest path, to implement relatively optimal load balancing. For the network device with either a ring network structure or a stereoscopic network structure, a requirement for a link bandwidth between the nodes in the network device can be reduced.

公开(公告)号：US20170109173A1

公开(公告)日：2017-04-20

申请号：US15391627

申请日：2016-12-27

Applicant: Huawei Technologies Co., Ltd.

Inventor： Nan LI ,  Linchun WANG

IPC: G06F9/445 ,  G06F9/44

CPC classification number: G06F9/445 ,  G06F8/30 ,  G06F8/61 ,  G06F8/65 ,  G06F9/44 ,  H04L41/5054 ,  H04L41/5077 ,  H04L45/38 ,  H04L49/70

Abstract: A method for determining instruction sequences, a related device, and a system. A method for determining instruction sequences may include: acquiring a function identifier; and determining M1 instruction sequences used to implement a function identified by the function identifier, where each instruction sequence in the M1 instruction sequences includes a unique entry instruction, and each instruction sequence in the M1 instruction sequences includes a unique exit instruction, where M1 is a positive integer, and the M1 instruction sequences are instruction sequences for creating a network packet processing program used to process a network packet. Solutions of embodiments of the present disclosure are advantageous for improving an extent of matching between a packet forwarding device and an instruction sequence that is executed by the packet forwarding device and used to process a network packet, and reducing an amount of invalid code run by the packet forwarding device.

公开(公告)号：US20230112747A1

公开(公告)日：2023-04-13

申请号：US17903227

申请日：2022-09-06

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Chao HUANG ,  Nan LI

IPC: H04L49/9005 ,  H04L47/70 ,  H04L49/90

Abstract: A method for allocating a resource for storing visualization information, an apparatus, and a system are provided. The method includes: a first network device determines a first queue based on a constraint condition, where the first queue is a queue that needs to be visualized. Then, the first network device allocates a first storage resource to the first queue, where the first storage resource is used to store visualization information of the first queue, and the visualization information is information used to visualize the first queue. Therefore, occupation of storage resources in the first network device is reduced.

公开(公告)号：US20220337505A1

公开(公告)日：2022-10-20

申请号：US17856790

申请日：2022-07-01

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yan ZHAO ,  Yibo YANG ,  Renjie QU ,  Nan LI

IPC: H04L45/02 ,  H04L45/00

Abstract: A method is applied to a ring link, where the ring link includes a first node, a second node, a third node, and a fourth node in sequence. According to the method, the first node receives first traffic, where the first node is a source node that sends the first traffic on the ring link; and the first node sends the first traffic to the third node, where two reachable paths with equal hop counts are included from the first node to the third node, the first node sends the first traffic to the third node on a preset first transmission path, the first transmission path passes through the second node, and the first transmission path is one of the two reachable paths with equal hop counts. This method can reduce computing load of nodes while implementing non-blocking switching of traffic between the nodes.