公开(公告)号：US20160187861A1

公开(公告)日：2016-06-30

申请号：US14585738

申请日：2014-12-30

Applicant: Futurewei Technologies Inc.

Inventor： Liya CHEN ,  Chen TIAN ,  Ziang HU

IPC: G05B13/02

CPC classification number: G05B13/027 ,  G06N99/005

Abstract: Methods and systems that facilitate efficient and effective adaptive execution mode selection are described. The adaptive execution mode selection is performed in part on-the-fly and changes to an execution mode (e.g., sequential, parallel, etc.) for a program task can be made. An intelligent adaptive selection can be made between a variety execution modes. The adaptive execution mode selection can also include selecting parameters associated with the execution modes. A controller receives historical information associated with execution mode selection, engages in training regarding execution mode selection, and adaptively selects an execution mode on-the-fly. The training can use an approach similar to an artificial neural network in which automated guided machine learning approach establishes correspondences between execution modes and task/input feature definitions based upon historical information. An adaptive selection is performed on-the-fly based on an initial trial run.

Abstract translation: 描述了促进有效和有效的自适应执行模式选择的方法和系统。 自适应执行模式选择部分地即时执行，并且可以进行对于程序任务的执行模式（例如，顺序，并行等）的改变。 可以在各种执行模式之间进行智能自适应选择。 自适应执行模式选择还可以包括选择与执行模式相关联的参数。 控制器接收与执行模式选择相关联的历史信息，参与有关执行模式选择的训练，并自动选择执行模式。 训练可以使用类似于人造神经网络的方法，其中自动引导机器学习方法基于历史信息建立执行模式与任务/输入特征定义之间的对应关系。 基于初始试运行，可以实时地进行自适应选择。

公开(公告)号：US20190007410A1

公开(公告)日：2019-01-03

申请号：US15640080

申请日：2017-06-30

Applicant: Futurewei Technologies, Inc.

Inventor： Luhui HU ,  Hui Zang ,  Ziang HU

IPC: H04L29/06 ,  H04L29/08 ,  H04L12/911 ,  H04L12/24 ,  H04L12/26 ,  H04L9/32

CPC classification number: H04L63/10 ,  H04L9/14 ,  H04L9/3297 ,  H04L41/147 ,  H04L41/20 ,  H04L41/5038 ,  H04L41/5096 ,  H04L43/0876 ,  H04L43/106 ,  H04L47/70 ,  H04L47/826 ,  H04L63/0281 ,  H04L63/166 ,  H04L67/10

Abstract: A system, computer readable medium, and method are provided for a resource management in a cloud architecture. The method includes the steps of collecting a first time stamped data (TSD), and a second TSD, and generating a prediction model based on the first TSD and the second TSD. The method further includes collecting a third TSD, and predicting a fourth TSD based on the prediction model and the third TSD. With more data are obtained via the prediction, the resource management is more efficient and accurate.