公开(公告)号：US20160112065A1

公开(公告)日：2016-04-21

申请号：US14984351

申请日：2015-12-30

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Deyuan CHANG ,  Zhiyu XIAO ,  Fan YU ,  Yu ZHAO

IPC: H03M13/11 ,  H03M13/00 ,  H03M13/37 ,  H04L25/03

CPC classification number: H03M13/1154 ,  H03M13/2957 ,  H03M13/2987 ,  H03M13/3746 ,  H03M13/3972 ,  H03M13/616 ,  H03M13/6331 ,  H03M13/6561 ,  H04B10/25073 ,  H04B10/2543 ,  H04B10/2569 ,  H04B10/6161 ,  H04B10/6162 ,  H04L1/0045 ,  H04L1/0057 ,  H04L25/0202 ,  H04L25/03057 ,  H04L25/03171 ,  H04L2025/03789

Abstract: Embodiments of the present application relate to a method for implementing Turbo equalization compensation. The equalizer divides a first data block into n data segments, where D bits in two adjacent data segments in the n data segments overlap, performs recursive processing on each data segment in the n data segments, before the recursive processing, merges the n data segments to obtain a second data block; and performs iterative decoding on the second data block, to output a third data block, where data lengths of the first data block, the second data block, and the third data block are all 1/T of a code length of a LDPC convolutional code.

Abstract translation: 本申请的实施例涉及一种用于实现Turbo均衡补偿的方法。 均衡器将第一数据块划分为n个数据段，其中n个数据段中的两个相邻数据段中的D位重叠，在递归处理之前对n个数据段中的每个数据段执行递归处理，合并n个数据段 以获得第二数据块; 对第二数据块执行迭代解码，输出第一数据块，第二数据块和第三数据块的数据长度都是LDPC卷积码的码长的1 / T的第三数据块 。

公开(公告)号：US20150160972A1

公开(公告)日：2015-06-11

申请号：US14557994

申请日：2014-12-02

Applicant: Huawei Technologies Co., Ltd.

Inventor： Fan YU ,  Yuwang Gong ,  Zhenguo Wang

IPC: G06F9/48

CPC classification number: G06F9/4856

Abstract: Embodiments of the present invention disclose a virtual machine migration management method, where the method includes: calculating, according to migration parameters of a to-be-migrated virtual machine, migration duration time required for migrating the to-be-migrated virtual machine from a source computing node to a destination computing node, where the migration parameters include an allocated memory size, a memory change rate, and migration network bandwidth that are of the to-be-migrated virtual machine; separately acquiring current available migration duration time of the source computing node and current available migration duration time of the destination computing node; and, if neither the current available migration duration time of the source computing node nor the current available migration duration time of the destination computing node is less than the migration duration time, determining to migrate the to-be-migrated virtual machine from the source computing node to the destination computing node.

Abstract translation: 本发明的实施例公开了一种虚拟机迁移管理方法，其中所述方法包括：根据待迁移的虚拟机的迁移参数，计算将迁移的虚拟机从 源计算节点到目的地计算节点，其中迁移参数包括被迁移的虚拟机的分配的存储器大小，存储器更改速率和迁移网络带宽; 分别获取源计算节点的当前可用迁移持续时间和目的地计算节点的当前可用迁移持续时间; 并且如果源计算节点的当前可用迁移持续时间和目的地计算节点的当前可用迁移持续时间都不小于迁移持续时间，则确定从源计算迁移待迁移虚拟机 节点到目标计算节点。

公开(公告)号：US20230087774A1

公开(公告)日：2023-03-23

申请号：US18071277

申请日：2022-11-29

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Ningxi NI ,  Min WANG ,  Zidong WANG ,  Fan YU ,  Lei CHEN

IPC: G06N3/08

Abstract: Example parameter optimization methods and apparatus are described. In one example parameter optimization method, a data processing device obtains a Kronecker factor matrix that is used to indicate a higher-order information matrix of a neural network model, and segments the Kronecker factor matrix to obtain a plurality of square matrices. The obtained plurality of square matrices are submatrices of the Kronecker factor matrix, and main diagonal lines of the plurality of square matrices each are in a one-to-one correspondence with a part of a main diagonal line of the Kronecker factor matrix. Then, the data processing device adjusts a parameter of the neural network model based on the obtained plurality of square matrices.

公开(公告)号：US20130174152A1

公开(公告)日：2013-07-04

申请号：US13621946

申请日：2012-09-18

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Fan YU

IPC: G06F9/455 ,  G01R21/00

CPC classification number: G01R21/00 ,  G06F9/455 ,  G06F9/45533 ,  G06F9/5077 ,  G06F9/5094 ,  G06F2009/4557 ,  Y02D10/22 ,  Y02D10/26 ,  Y02D10/28 ,  Y02D10/36

Abstract: Embodiments of the present invention relate to a virtual machine integration technology, and in particular, to a method, an apparatus, and a system for virtual cluster integration. The method includes: performing a calculation through a search algorithm to obtain the minimum number of physical machines which are capable of accommodating all virtual machines in a virtual cluster, and obtaining all virtual integration solutions satisfying the minimum number of physical machines; then calculating CPU voltage consumption of each virtual integration solution, and selecting a solution with lowest CPU voltage consumption from these virtual integration solutions; and formulating a virtual integration migration policy according to the virtual integration solution with the lowest CPU voltage consumption. Therefore, through the embodiments of the present invention, a virtual integration solution with lower CPU voltage energy consumption can be obtained, thereby greatly improving an energy saving and emission reduction effect of a virtual cluster integration solution.

Abstract translation: 本发明的实施例涉及一种虚拟机集成技术，特别涉及用于虚拟集群集成的方法，装置和系统。 该方法包括：通过搜索算法执行计算，以获得能够容纳虚拟集群中的所有虚拟机的物理机的最小数量，并获得满足最小数量的物理机的所有虚拟集成解; 然后计算每个虚拟集成解决方案的CPU电压消耗，并从这些虚拟集成解决方案中选择具有最低CPU电压消耗的解决方案; 并根据具有最低CPU电压消耗的虚拟集成解决方案制定虚拟集成迁移策略。 因此，通过本发明的实施例，可以获得具有较低CPU电压能量消耗的虚拟集成解决方案，从而大大提高了虚拟集群集成解决方案的节能和减排效果。

公开(公告)号：US20240005541A1

公开(公告)日：2024-01-04

申请号：US18468293

申请日：2023-09-15

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yue GONG ,  Xinjun MA ,  Jianan HE ,  LEI CHEN ,  Fan YU

IPC: G06T7/593 ,  G06T3/40 ,  G06T7/62 ,  G06V10/44

CPC classification number: G06T7/593 ,  G06T3/40 ,  G06T7/62 ,  G06V10/44

Abstract: This application relates to the field of image processing technologies, and in some embodiments, to an image depth prediction method and an electronic device. The method includes: obtaining a primary view and a first secondary view, where the primary view includes a first pixel, the first pixel corresponds to a second pixel at a first assumed depth, and the second pixel is located on the first secondary view; updating feature information of the second pixel based on the feature information of the second pixel and feature information of at least one third pixel, the fourth pixel is a pixel that corresponds to the first pixel at a second assumed depth and that is on the first secondary view; and obtaining a probability of the first assumed depth based on the feature information of the first pixel and the updated feature information of the second pixel.