公开(公告)号：US20160277292A1

公开(公告)日：2016-09-22

申请号：US15166072

申请日：2016-05-26

Applicant: Huawei Technologies Co., Ltd.

Inventor： Xinyu HU ,  Yuping ZHAO

IPC: H04L12/713 ,  H04L12/931 ,  H04L29/08 ,  H04L12/24

CPC classification number: H04L45/586 ,  G06F9/5077 ,  H04L41/0803 ,  H04L41/0816 ,  H04L49/70 ,  H04L67/1006 ,  H04L67/32 ,  H04L69/12

Abstract: A network service processing method and apparatus are provided. The method includes: partitioning a field-programmable gate array (FPGA) or a network processor (NP) into multiple mutually isolated service processing resources; receiving a first configuration instruction, where the first configuration instruction includes a first service configuration execution file; performing service configuration on a service processing resource according to the first service configuration execution file, so that the service processing resource performs service processing on a to-be-processed service packet; receiving a forwarding rule used when service forwarding is to be performed on the to-be-processed service packet; and receiving the to-be-processed service packet sent by a user terminal, and distributing the to-be-processed service packet to the service processing resource by using the forwarding rule, so that the service processing resource performs service processing on the to-be-processed service packet.

Abstract translation: 提供一种网络服务处理方法和装置。 该方法包括：将现场可编程门阵列（FPGA）或网络处理器（NP）分成多个相互隔离的服务处理资源; 接收第一配置指令，其中所述第一配置指令包括第一服务配置执行文件; 根据所述第一服务配置执行文件在服务处理资源上执行服务配置，使得所述服务处理资源对待处理的服务分组执行服务处理; 接收在对待处理的业务分组进行业务转发时使用的转发规则; 接收用户终端发送的待处理业务报文，并通过转发规则将待处理业务报文分发给业务处理资源，使业务处理资源对业务处理资源进行业务处理， 被处理的服务包。

公开(公告)号：US20150127849A1

公开(公告)日：2015-05-07

申请号：US14593577

申请日：2015-01-09

Applicant: Huawei Technologies Co., Ltd.

Inventor： Hai LUO ,  Xinyu HU ,  Qikun WEI ,  Liang ZHANG

IPC: H04L29/06 ,  H04L12/801

CPC classification number: H04L69/16 ,  H04L47/12 ,  H04L49/9063 ,  H04L69/12 ,  H04L69/161 ,  H04L69/163

Abstract: Embodiments of the present invention provide a transmission control protocol TCP data transmission method, a TCP offload engine, and a system, which relate to the field of communications, and can reduce data migration between the TCP offload engine and a CPU, and at the same time reduce parsing work on data by the CPU, so as to achieve the effects of reducing resources of the CPU for processing TCP/IP data and reducing transmission delay. The method includes: a TCP offload engine receives TCP data sent by a remote device; performs TCP offloading on the TCP data; identifies the TCP offloaded data, and sends, according to an identification result, the TCP offloaded data to a CPU or a storage device corresponding to storage position information issued by the CPU. The embodiments of the present invention are applicable to TCP data transmission.

Abstract translation: 本发明的实施例提供了一种与通信领域相关的传输控制协议TCP数据传输方法，TCP卸载引擎和系统，并且可以减少TCP卸载引擎和CPU之间的数据迁移，并且同时 时间减少了CPU对数据的解析工作，从而实现了减少CPU资源处理TCP / IP数据和减少传输延迟的效果。 该方法包括：TCP卸载引擎接收远程设备发送的TCP数据; 对TCP数据执行TCP卸载; 识别TCP卸载的数据，并根据识别结果将TCP卸载的数据发送到与由CPU发布的存储位置信息相对应的CPU或存储设备。 本发明的实施例可应用于TCP数据传输。

公开(公告)号：US20220366310A1

公开(公告)日：2022-11-17

申请号：US17871084

申请日：2022-07-22

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Xiaoyun SI ,  Liang ZHANG ,  Xinyu HU ,  Jun WU

IPC: G06N20/00

Abstract: A method for implementing model update is provided and used in a federated update framework. The method includes: a service device receives difference information uploaded by at least two clients. The service device performs calculation based on the difference information uploaded by the at least two clients, to obtain first difference consistency information, where the first difference consistency information indicates a consistency degree of the difference information uploaded by the at least two clients. The service device performs calculation based on the first difference consistency information, to obtain first training information, where the first training information is used to train a third model, and the third model is obtained by updating the first model by the service device based on the difference information uploaded by the at least two clients. The service device sends the first training information to the at least two clients.