公开(公告)号：US20240428135A1

公开(公告)日：2024-12-26

申请号：US18747830

申请日：2024-06-19

Applicant: Hitachi, Ltd.

Inventor： Tsutomu KUSABA ,  Soichi TAKASHIGE ,  Daisuke KOMAKI

IPC: G06N20/00

Abstract: A machine learning development support system includes: a data storage unit that stores training data; a data attribute extraction unit that refers to definition information whereby a conversion condition for converting the training data into an attribute thereof is defined and extracts data acquisition conditions from the training data; and an experiment record information storage unit that stores experiment record information., The machine learning product including a machine learning program and a trained model is divided into versions and recorded, and an accuracy of an inference obtained by each version is recorded in association with the version, the trained model, and the training data; and a processing specification unit that refers to the experiment record information, and specifies, for each of the data acquisition conditions, a machine learning product in which the accuracy of the inference is improved between old and new versions of the machine learning product.

公开(公告)号：US20240020577A1

公开(公告)日：2024-01-18

申请号：US18216647

申请日：2023-06-30

Applicant: Hitachi, Ltd.

Inventor： Itsumi TSUCHIYA ,  Soichi TAKASHIGE ,  Tatsuhiro MATSUI

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: In a data management system of a machine learning model, flag management information (a flag importance management table) manages and defines respective flags corresponding to, of a plurality of processes included in the life cycle, one or more predetermined processes. An operation unit assigns flags defined in the flag management information to input data and output data of the model in accordance with involvement in the predetermined processes when the model is operated. A data management unit determines, with respect to each of the input data and the output data, the necessity of storage of data on the basis of a flag assigned to the data by the operation unit.

公开(公告)号：US20210357809A1

公开(公告)日：2021-11-18

申请号：US17191813

申请日：2021-03-04

Applicant: HITACHI, LTD.

Inventor： Kyosuke HASHIMOTO ,  Soichi TAKASHIGE

IPC: G06N20/00 ,  G06K9/62

Abstract: To efficiently and appropriately improve an inference model when the inference model outputs an inappropriate inference result that affects work. A model improvement system performs inference related to the work using an inference model which is a machine learning model that outputs an inference result by inputting an input matrix, acquires an actual result which is information acquired by actually performing work to be inferred by the inference model, determines whether or not the inference result is an inappropriate content that affects the work by comparing the inference result with the actual result, divides a part of the input matrix that outputs the inference result that affects the work from the input matrix, and generates an inference model that outputs an inference result that does not affect the work when the part divided from the input matrix is input as an input matrix.

公开(公告)号：US20220365526A1

公开(公告)日：2022-11-17

申请号：US17691784

申请日：2022-03-10

Applicant: Hitachi, Ltd.

Inventor： Sayuri ISHIKAWA ,  Soichi TAKASHIGE ,  Masafumi TSUYUKI ,  Daisuke KOMAKI

IPC: G05B23/02

Abstract: When an event such as maintenance occurs, a model in which a tendency of input data changes due to an influence of the maintenance and retraining is required is specified. Model identifiers that specify the machine learning models and are unique to the machine learning models and sensor identifiers that specify the sensors that output the sensor data serving as the input data of the machine learning models and are unique to the sensors are managed to be in association with each other. A degree of influence indicating a change in tendency of the sensor data before and after the maintenance operation is performed is obtained for each maintenance event identifier that specifies a maintenance operation performed on a device and is unique to the maintenance operation, and the degree of influence is managed in association with each of the sensor identifiers. When the sensor whose degree of influence satisfies a predetermined condition is influenced by the maintenance operation, the model identifier associated with the sensor identifier of the sensor satisfying the condition is presented.

公开(公告)号：US20200065316A1

公开(公告)日：2020-02-27

申请号：US16288490

申请日：2019-02-28

Applicant: Hitachi, Ltd.

Inventor： Jun MIZUNO ,  Takashi TAMESHIGE ,  Soichi TAKASHIGE ,  Nobuhiro MAKI

IPC: G06F16/27 ,  G06F3/06 ,  H04L29/08

Abstract: The present invention ensures improving responsiveness in a P2P type distributed database system that includes a plurality of node computers capable of storing management target data in storage devices. The storage devices include a first type storage device having a high input/output performance and a second type storage device having a low input/output performance. There is provided the storage devices that store KeySpace management tables including response count information indicative of a count of the node computers from which the identical management target data are to be read. The identical management target data are required to respond to a read request of the management target data. The node computers by a count indicated by the response count information and capable of storing the management target data in the first type storage devices among the plurality of node computers are determined as storage destinations when a CPU receives a write request of new management target data. The determined node computers are caused to store the management target data.

公开(公告)号：US20190332275A1

公开(公告)日：2019-10-31

申请号：US16299242

申请日：2019-03-12

Applicant: HITACHI, LTD.

Inventor： Azusa JIN ,  Hideo SAITO ,  Masakuni AGETSUMA ,  Soichi TAKASHIGE

IPC: G06F3/06 ,  G06F11/14 ,  G06F16/182 ,  G06F11/20

Abstract: The present invention proposes an information processing system and a volume allocation method making it feasible to reduce overall system operation cost. A replication group to carry out replication is comprised of one or multiple nodes. A storage node classifies respective storage devices mounted thereon into plural storage hierarchies according to the drive type of each of the storage devices for management of the storage devices. A controller node allocates, to each node of the nodes constituting the replication group, a volume for which a storage device of a storage hierarchy suitable for a type of middleware that the node uses and a node type of the node acting in the replication group provides a storage area.

公开(公告)号：US20210048869A1

公开(公告)日：2021-02-18

申请号：US16794287

申请日：2020-02-19

Applicant: HITACHI, LTD.

Inventor： Jun MIZUNO ,  Soichi TAKASHIGE ,  Keisuke HATASAKI

IPC: G06F1/3212 ,  G06F1/3234 ,  G06F1/3228 ,  H04W52/02

Abstract: Status information including remaining battery power is collected from a management target device. Subsequent operations of the device are identified from an operation schedule management collection. Battery consumptions corresponding to subsequent operations of the device are estimated from KPI management collection that includes information representing relationships between battery consumptions and operations with respect to each of a plurality of devices. The battery life of the device is predicted based on the estimated battery consumptions and the remaining battery power represented by the collected status information. A circuit schedule of a maintenance engineer is determined based on the predicted battery life, the schedule including necessity of dispatch of the maintenance engineer to the device.

公开(公告)号：US20190250994A1

公开(公告)日：2019-08-15

申请号：US16119087

申请日：2018-08-31

Applicant: HITACHI, LTD.

Inventor： Hideo SAITO ,  Takaki NAKAMURA ,  Soichi TAKASHIGE ,  Masakuni AGETSUMA

IPC: G06F11/14 ,  G06F9/455

CPC classification number: G06F11/1458 ,  G06F9/45558 ,  G06F2009/4557 ,  G06F2009/45579 ,  G06F2009/45583 ,  G06F2201/815

Abstract: To reduce cumbersomeness in manually setting backup requirements per middleware. Not only one or more deployment requirements but also one or more first backup requirements are associated with a middleware deployment indication. In response to the indication, middleware deployment and backup setting are performed. The middleware deployment includes associating volumes (VOLs) the number of which complies with the one or more deployment requirements, with virtual machines (VMs) the number of which complies with the one or more deployment requirements. The backup setting includes associating two or more backup requirements based on the one or more first backup requirements and one or more predetermined second backup requirements in response to a middleware type of middleware to be deployed, with the middleware deployment.

公开(公告)号：US20130283261A1

公开(公告)日：2013-10-24

申请号：US13914631

申请日：2013-06-11

Applicant: HITACHI, LTD.

Inventor： Hideki TAKANO ,  Soichi TAKASHIGE

IPC: G06F9/445

CPC classification number: G06F9/5011 ,  G06F8/61 ,  G06F8/62 ,  G06F9/5055 ,  H04L67/1002 ,  H04L67/1008 ,  H04L67/1012

Abstract: Standby computers are dynamically divided into groups according to the difference between the software thereof and the software required of a business system. When a computer is made available, the standby computers divided into groups by software structure are searched and an appropriate one is extracted to quickly complete the construction of the software environment. An active computer, if to be transferred to standby mode, is associated with the group having the same software structure as the business system with which the active computer has thus far operated. The active/standby states of the computers are monitored, and the standby computers are changed thereby to widen the possible range of application to other business systems, removing copy operation of disk images from the active computer to the standby computers.

Abstract translation: 备用计算机根据其软件与业务系统所需的软件之间的差异动态分为组。 当计算机可用时，搜索通过软件结构分组的备用计算机，并提取适当的计算机，以快速完成软件环境的构建。 如果要转移到待机模式，活动计算机与具有与活动计算机迄今为止运行的业务系统具有相同软件结构的组相关联。 监视计算机的主动/待机状态，并且更改备用计算机，从而扩大对其他业务系统的可能应用范围，从而将磁盘映像从活动计算机复制到备用计算机。

公开(公告)号：US20210365813A1

公开(公告)日：2021-11-25

申请号：US17209341

申请日：2021-03-23

Applicant: Hitachi, Ltd.

Inventor： Kaori NAKANO ,  Masaharu UKEDA ,  Soichi TAKASHIGE ,  Yuxin LIANG

IPC: G06N5/04 ,  G06N20/00

Abstract: A management computer for managing a system that makes an inference using a training model has a processor for performing a process in cooperation with a memory, and the processor executes: a generation process for generating an accuracy improvement prediction model for predicting the accuracy of a retrained model when retraining is executed using retraining data including new collected data collected from the system after the start of the operation of the system based on a correlation between the Feature of training data used for training of the training model and the accuracy of the training model; a prediction process for predicting the accuracy of the retrained model from the accuracy improvement prediction model and the Feature of the retraining data; and a determination process for determining whether or not the execution of the retraining is necessary based on the predicted accuracy of the retrained model.