-
公开(公告)号：US12206528B2
公开(公告)日：2025-01-21
申请号：US18333760
申请日：2023-06-13
Applicant: Huawei Technologies Co., Ltd.
Inventor： Dongdong Shen , Zhaohua Qian , Jingjing Wang , Guang Qian , Xiaowei Li
Abstract: A communication system includes a first device, a second device, and a cable that connects the first device and the second device, and the cable stores cable information of the cable. A link training method includes: obtaining device information of the first device, obtaining the cable information from the cable, and obtaining device information of the second device; and performing link training between the first device and the second device based on the device information of the first device, the cable information, and the device information of the second device.
-
公开(公告)号：US20230379076A1
公开(公告)日：2023-11-23
申请号：US18361409
申请日：2023-07-28
Applicant: Huawei Technologies Co., Ltd.
Inventor： Xiaowei Li , Zhenxing Zhang , Junhao Jin , Guang Qian , Jingjing Wang
IPC: H04L1/00
CPC classification number: H04L1/0002
Abstract: A data transmission method and apparatus are provided. The method includes: obtaining service data; and sending first data, where the first data is obtained based on the service data, where a data structure of the first data is the same as a data structure of second data, the second data is for performing link training, and the link training is performed before the first data is sent.
-
公开(公告)号：US09965335B2
公开(公告)日：2018-05-08
申请号：US14940577
申请日：2015-11-13
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hang Lu , Yinhe Han , Binzhang Fu , Xiaowei Li
CPC classification number: G06F9/5088 , G06F9/5066 , G06F15/7825
Abstract: A task allocation method, a chip are disclosed. The method includes: determining a number of threads included in a to-be-processed task; determining, in a network-on-chip formed by a multi-core processor, a continuous area formed by routers-on-chip corresponding to multiple continuous idle processor cores whose number is equal to the number of the threads; if the area is a non-rectangular area, determining a rectangular area extended from the area; and if predicted traffic of each router-on-chip that is connected to a non-idle processor core and in the extended rectangular area does not exceed a preset threshold, allocating the multiple threads of the to-be-processed task to the idle processor cores in the area. According to the task allocation method provided in the embodiments of the present invention, problems of large hardware overheads, a low network throughput, low system utilization are avoided.
-
公开(公告)号：US20230344683A1
公开(公告)日：2023-10-26
申请号：US18333760
申请日：2023-06-13
Applicant: Huawei Technologies Co., Ltd.
Inventor： Dongdong Shen , Zhaohua Qian , Jingjing Wang , Guang Qian , Xiaowei Li
IPC: H04L25/03
CPC classification number: H04L25/03012
Abstract: A communication system includes a first device, a second device, and a cable that connects the first device and the second device, and the cable stores cable information of the cable. A link training method includes: obtaining device information of the first device, obtaining the cable information from the cable, and obtaining device information of the second device; and performing link training between the first device and the second device based on the device information of the first device, the cable information, and the device information of the second device.
-
Patent Agency Ranking
公开(公告)号：US20160226633A1
公开(公告)日：2016-08-04
申请号：US15096393
申请日：2016-04-12
Applicant: Huawei Technologies Co., Ltd.
Inventor： Xing Hu , Yu Hu , Shuchang Shan , Xiaowei Li
CPC classification number: H04L1/22 , H04L1/0007 , H04L29/12575 , H04L45/245 , H04L45/28 , H04L47/10 , H04L69/324 , H04L69/326 , H04L2001/0094
Abstract: A data transmission method and apparatus, where the method comprises checking full-bandwidth transmission paths of a bus, and When a fault occurs in the full-bandwidth transmission paths and a quantity of faulty full-bandwidth transmission paths is less than or equal to M, selecting N full-bandwidth transmission paths from full-bandwidth transmission paths that are not faulty to transmit a data unit, and when a fault occurs in the full-bandwidth transmission paths and a quantity of faulty full-bandwidth transmission paths is greater than M, reconfiguring a size of a data unit according to a quantity of full-bandwidth transmission paths that are not faulty and a target burst quantity.
Abstract translation: 一种数据传输方法和装置，其中该方法包括检查总线的全带宽传输路径，当在全带宽传输路径中出现故障时，一批有缺陷的全带宽传输路径小于或等于M 从全带宽传输路径中选择N个全带宽传输路径，这些传输路径不发送数据单元，当全带宽传输路径发生故障时，一批故障全带宽传输路径大于M 根据不具有故障的全带宽传输路径的数量和目标突发量重新配置数据单元的大小。
-
6.发明授权公开(公告)号：US10671447B2
公开(公告)日：2020-06-02
申请号：US15943370
申请日：2018-04-02
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hang Lu , Yinhe Han , Binzhang Fu , Xiaowei Li
Abstract: A task allocation method, a chip are disclosed. The method includes: determining the number of threads included in a to-be-processed task; determining, in a network-on-chip formed by a multi-core processor, a continuous area formed by routers-on-chip corresponding to multiple continuous idle processor cores whose number is equal to the number of the threads; when the area is a non-rectangular area, determining an extended area extended from the non-rectangular area; and when predicted traffic of each router-on-chip that is connected to a processor core in the extended area does not exceed a preset threshold, allocating the multiple threads of the to-be-processed task to the idle processor cores in the non-rectangular area. According to the task allocation method provided in the embodiments of the present invention, problems of large hardware overheads, a low network throughput, low system utilization are avoided.
-
公开(公告)号：US10069604B2
公开(公告)日：2018-09-04
申请号：US15096393
申请日：2016-04-12
Applicant: Huawei Technologies Co., Ltd.
Inventor： Xing Hu , Yu Hu , Shuchang Shan , Xiaowei Li
IPC: H04L1/22 , H04L12/801 , H04L12/709 , H04L29/08 , H04L29/12 , H04L12/703 , H04L1/00
Abstract: A data transmission method and apparatus, where the method comprises checking full-bandwidth transmission paths of a bus, and When a fault occurs in the full-bandwidth transmission paths and a quantity of faulty full-bandwidth transmission paths is less than or equal to M, selecting N full-bandwidth transmission paths from full-bandwidth transmission paths that are not faulty to transmit a data unit, and when a fault occurs in the full-bandwidth transmission paths and a quantity of faulty full-bandwidth transmission paths is greater than M, reconfiguring a size of a data unit according to a quantity of full-bandwidth transmission paths that are not faulty and a target burst quantity.
-
8.发明申请公开(公告)号：US20160070603A1
公开(公告)日：2016-03-10
申请号：US14940577
申请日：2015-11-13
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hang Lu , Yinhe Han , Binzhang Fu , Xiaowei Li
CPC classification number: G06F9/5088 , G06F9/5066 , G06F15/7825
Abstract: A task allocation method, a chip are disclosed. The method includes: determining a number of threads included in a to-be-processed task; determining, in a network-on-chip formed by a multi-core processor, a continuous area formed by routers-on-chip corresponding to multiple continuous idle processor cores whose number is equal to the number of the threads; if the area is a non-rectangular area, determining a rectangular area extended from the area; and if predicted traffic of each router-on-chip that is connected to a non-idle processor core and in the extended rectangular area does not exceed a preset threshold, allocating the multiple threads of the to-be-processed task to the idle processor cores in the area. According to the task allocation method provided in the embodiments of the present invention, problems of large hardware overheads, a low network throughput, low system utilization are avoided.
Abstract translation: 公开了一种任务分配方法，芯片。 该方法包括：确定包括在待处理任务中的线程数; 在由多核处理器形成的芯片上的片上网络中，确定由与所述线程数相等的多个连续空闲处理器核心对应的片上路由器形成的连续区域; 如果该区域是非矩形区域，则确定从该区域延伸的矩形区域; 并且如果连接到非空闲处理器核心并且在扩展矩形区域中的每个片上路由器的预测流量不超过预设阈值，则将待处理任务的多个线程分配给空闲处理器 核心在该地区。 根据本发明实施例提供的任务分配方法，避免了大的硬件开销，低网络吞吐量，低系统利用率的问题。
-
公开(公告)号：US20180225156A1
公开(公告)日：2018-08-09
申请号：US15943370
申请日：2018-04-02
Applicant: Huawei Technologies Co., Ltd.
Inventor： Hang Lu , Yinhe Han , Binzhang Fu , Xiaowei Li
Abstract: A task allocation method, a chip are disclosed. The method includes: determining the number of threads included in a to-be-processed task; determining, in a network-on-chip formed by a multi-core processor, a continuous area formed by routers-on-chip corresponding to multiple continuous idle processor cores whose number is equal to the number of the threads; when the area is a non-rectangular area, determining an extended area extended from the non-rectangular area; and when predicted traffic of each router-on-chip that is connected to a processor core in the extended area does not exceed a preset threshold, allocating the multiple threads of the to-be-processed task to the idle processor cores in the non-rectangular area. According to the task allocation method provided in the embodiments of the present invention, problems of large hardware overheads, a low network throughput, low system utilization are avoided.
-
-
-
-
-
-
-
-
Search Results
9
records
(took 0.043 seconds)
