WO2022267085A1 - Artificial-intelligence-based data management method and system for data center - Google Patents

Artificial-intelligence-based data management method and system for data center Download PDF

Info

Publication number
WO2022267085A1
WO2022267085A1 PCT/CN2021/103347 CN2021103347W WO2022267085A1 WO 2022267085 A1 WO2022267085 A1 WO 2022267085A1 CN 2021103347 W CN2021103347 W CN 2021103347W WO 2022267085 A1 WO2022267085 A1 WO 2022267085A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
weight
historical
scheduling
priority
Prior art date
Application number
PCT/CN2021/103347
Other languages
French (fr)
Chinese (zh)
Inventor
徐一忠
陈晗
何水芳
徐政宇
Original Assignee
浙江海瑞网络科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 浙江海瑞网络科技有限公司 filed Critical 浙江海瑞网络科技有限公司
Publication of WO2022267085A1 publication Critical patent/WO2022267085A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/906—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the present invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based data center data management method and system.
  • embodiments of the present invention provide an artificial intelligence-based data center data management method and system.
  • An embodiment of the present invention provides an artificial intelligence-based data center data management method, including:
  • the data attributes include data source, data type, data size, data purpose label, and the scheduling records Including data migration, data deletion, data merging;
  • the method also includes:
  • a comprehensive weight is determined according to the first weight, the second weight, and the third weight, and a corresponding data attribute weight model is constructed according to the remaining historical data and the corresponding comprehensive weight.
  • the method also includes:
  • the method also includes:
  • the method also includes:
  • the data attributes of the target data are respectively input to the three training convolutional neural network models, and one main scheduling record and two additional records are output.
  • the method also includes:
  • the training sample is divided into a training set and a verification set, and the training set is input to a convolutional neural network model for training to obtain a trained preliminary convolutional neural network model;
  • the verification set is input to the trained preliminary convolutional neural network model for testing, and the trained convolutional neural network model is obtained after the test is completed.
  • An embodiment of the present invention provides an artificial intelligence-based data center data management system, including:
  • An acquisition module configured to acquire historical scheduling data in historical records, and acquire data attributes and scheduling records of corresponding historical data according to the historical scheduling data, the data attributes including data sources, data types, data sizes, and data purpose tags , the scheduling record includes data migration, data deletion, and data merging;
  • An exclusion module configured to exclude data containing data purpose tags in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
  • the input module is used to input the data attribute weight model and the corresponding scheduling record into the convolutional neural network model as training samples for training;
  • the output module is used to obtain the data attributes of the target data, and input the data attributes of the target data to the trained convolutional neural network model, output the scheduling records corresponding to the target data, and perform the scheduling according to the scheduling records of the target data.
  • the above target data is used for data management.
  • the system also includes:
  • a first determining module configured to determine a first priority of the sender according to the data source, and determine a first weight of the remaining historical data according to the first priority
  • a second determining module configured to determine a second priority of the remaining historical data according to the data type, and determine a second weight of the remaining historical data according to the second priority
  • a third determining module configured to determine a third priority of the remaining historical data according to the data size, and determine a third weight of the remaining historical data according to the third priority
  • a construction module configured to determine a comprehensive weight according to the first weight, the second weight, and the third weight, and construct a corresponding data attribute weight model according to the remaining historical data and the corresponding comprehensive weight.
  • An embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the program, the above-mentioned artificial intelligence-based data center data management is realized. method steps.
  • An embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned artificial intelligence-based data center data management method are implemented.
  • the artificial intelligence-based data center data management method and system acquire historical scheduling data in historical records, and acquire data attributes and scheduling records of corresponding historical data according to the historical scheduling data.
  • the data attributes Including data sources, data types, data sizes, and data purpose tags, the scheduling records include data migration, data deletion, and data merging; data containing data purpose tags in the historical data are excluded, and according to the data sources and data of the remaining historical data Type and data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and
  • the data attributes are input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and the target data is managed according to the scheduling record of the target data.
  • the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
  • Fig. 1 is the flow chart of the data center data management method based on artificial intelligence in the embodiment of the present invention
  • Fig. 2 is the structural diagram of the data center data management system based on artificial intelligence in the embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of an electronic device in an embodiment of the present invention.
  • Fig. 1 is a schematic flowchart of an artificial intelligence-based data center data management method provided by an embodiment of the present invention. As shown in Fig. 1 , the embodiment of the present invention provides an artificial intelligence-based data center data management method, including:
  • Step S101 obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data
  • the data attributes include data source, data type, data size, data destination tag, all
  • the above scheduling records include data migration, data deletion, and data merging.
  • the historical records in the data center are acquired, wherein the historical scheduling data in the historical records is acquired, and the historical scheduling data are the data attributes of the historical data itself and the corresponding data scheduling records, wherein the data attributes may include data sources, data types , data size, data destination label (the label of the data sending destination), and scheduling records include data migration, data deletion, and data merging.
  • Step S102 excluding data containing data purpose tags from the historical data, and constructing a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data.
  • the above data is excluded, and then data management is performed on the data without a clear data processing destination, through the remaining
  • the data source, data type, and data size of historical data are input, and the corresponding data attribute weight model is constructed.
  • Step S103 input the data attribute weight model and the corresponding scheduling records as training samples into the convolutional neural network model for training.
  • the data attribute weight model constructed according to the remaining historical data and the scheduling records corresponding to the remaining historical data are used as input, and input to the input layer of the convolutional neural network model for model training.
  • the convolutional neural network model passes through the convolutional layer-pool Deep learning is performed on the layer-full connection layer to obtain the trained convolutional neural network model.
  • Step S104 Obtain the data attribute of the target data, input the data attribute of the target data into the trained convolutional neural network model, output the scheduling record corresponding to the target data, and process the target according to the scheduling record of the target data Data for data management.
  • the data attributes of the target data that need to be managed are obtained, and the data attributes of the target data are input into the trained convolutional neural network model, and the scheduling records corresponding to the target data are output.
  • the scheduling records can be for the target data. Data migration, data deletion, data merging, and data management of the target data according to the data scheduling records.
  • the scheduling records of the data attribute weight model and data migration, data deletion, and data merging can also be input into the convolutional neural network model as training samples for training, and three training convolutional neural network models are obtained respectively, so that Input the data attributes of the target data into three training convolutional neural network models respectively, and output 1 main scheduling record and 2 additional records.
  • the management method of the main scheduling record being recorded in the three records is more complete and more in line with the data
  • the required record is the main data management method
  • the 2 additional records are the secondary data management method of the target data, which does not require high integrity of the management method of the record, thereby improving the diversity of data management.
  • An artificial intelligence-based data center data management method acquires historical scheduling data in historical records, and acquires data attributes and scheduling records of corresponding historical data according to the historical scheduling data.
  • the data attributes Including data sources, data types, data sizes, and data purpose tags, the scheduling records include data migration, data deletion, and data merging; data containing data purpose tags in the historical data are excluded, and according to the data sources and data of the remaining historical data Type and data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and
  • the data attributes are input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and the target data is managed according to the scheduling record of the target data.
  • the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
  • the artificial intelligence-based data center data management method further includes:
  • a comprehensive weight is determined according to the first weight, the second weight, and the third weight, and a corresponding data attribute weight model is constructed according to the remaining historical data and the corresponding comprehensive weight.
  • the first priority of the sender is determined according to the source of the data, that is, the first priority is determined according to the priority of the sender (such as the importance of the customer), and then the history
  • the first weight of the data and then determine the second priority of the historical data according to the data type, for example, the higher the processing priority of the data type, the higher the second priority, and then determine the second weight of the historical data, according to
  • the data size determines the third priority of historical data.
  • the third priority can be determined by obtaining the current capacity of the data center, according to the size of the data, combined with the current capacity, and dynamically adjusting the third priority of the remaining historical data.
  • the weight tendency includes giving priority to the sender (that is, increasing the first weight, reducing the second weight, and the third weight,
  • the initial weights of the first weight, the second weight, and the third weight are all 1/3, and the first weight is increased, the second weight is decreased, and the third weight is increased according to the degree of weight tendency, and the priority data is given priority (that is, increase second weight, reduce the first weight, third weight), data center performance priority (that is, increase the third weight, reduce the first weight, second weight) one of the three, and according to the data source of the data in the remaining historical data,
  • the corresponding data attribute weight model is constructed through the three weights of the remaining historical data, and then the deep learning management of artificial intelligence is completed through the data attribute weight model.
  • the artificial intelligence-based data center data management method further includes:
  • the training sample is divided into a training set and a verification set, and the training set is input to a convolutional neural network model for training to obtain a trained preliminary convolutional neural network model;
  • the verification set is input to the trained preliminary convolutional neural network model for testing, and the trained convolutional neural network model is obtained after the test is completed.
  • a part of sample data is usually separated as a verification set to verify the data accuracy of the trained convolutional neural network model.
  • the training samples of the data attribute weight model and the corresponding scheduling records are grouped. Specifically, the sample data can be divided into 80% of the training set and 20% of the verification set.
  • Preliminary training is performed on the data attribute weight model and scheduling records corresponding to the training set data to obtain the preliminary convolutional neural network model, and then the preliminary convolutional neural network model is trained through the verification set and the data attribute weight model and scheduling records corresponding to the verification set data. Test to get the trained convolutional neural network model.
  • the accuracy of the convolutional neural network model is ensured by grouping the sample data into data groups, establishing a preliminary model through the training set, and verifying the accuracy of the preliminary model through the verification set.
  • Fig. 2 is an artificial intelligence-based metering management system provided by an embodiment of the present invention, including: an acquisition module S201, an exclusion module S202, an input module S203, and an output module S204, wherein:
  • the obtaining module S201 is used to obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, and the data attributes include data source, data type, data size, and data purpose label, the scheduling records include data migration, data deletion, and data merging.
  • the exclusion module S202 is configured to exclude data containing data purpose tags from the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data.
  • the input module S203 is used for inputting the data attribute weight model and the corresponding scheduling records as training samples into the convolutional neural network model for training.
  • the output module S204 is used to obtain the data attributes of the target data, and input the data attributes of the target data into the trained convolutional neural network model, and output the scheduling records corresponding to the target data, according to the scheduling records of the target data
  • the target data is subject to data management.
  • system may also include:
  • a first determining module configured to determine a first priority of the sender according to the data source, and determine a first weight of the remaining historical data according to the first priority.
  • the second determining module is configured to determine a second priority of the remaining historical data according to the data type, and determine a second weight of the remaining historical data according to the second priority.
  • a third determining module configured to determine a third priority of the remaining historical data according to the data size, and determine a third weight of the remaining historical data according to the third priority.
  • a construction module configured to determine a comprehensive weight according to the first weight, the second weight, and the third weight, and construct a corresponding data attribute weight model according to the remaining historical data and the corresponding comprehensive weight.
  • system may also include:
  • the second acquisition module is configured to acquire the current capacity of the data center, and dynamically adjust the third priority of the remaining historical data according to the size of the data and in combination with the current capacity.
  • system may also include:
  • the third acquisition module is used to obtain a preset weight tendency, and according to the weight tendency, combined with the first weight, the second weight, and the third weight to determine a comprehensive weight, the weight tendency is sender priority, priority data One of priority and data center performance priority.
  • system may also include:
  • the training module is used to divide the training samples into a training set and a verification set, input the training set into the convolutional neural network model for training, and obtain a trained preliminary convolutional neural network model.
  • the test module is used for inputting the verification set into the trained preliminary convolutional neural network model for testing, and obtaining the trained convolutional neural network model after the test is completed.
  • Each module in the above artificial intelligence-based data center data management system can be realized in whole or in part by software, hardware and a combination thereof.
  • the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.
  • FIG. 3 illustrates a schematic diagram of the physical structure of an electronic device.
  • the electronic device may include: a processor (processor) 301, a memory (memory) 302, a communication interface (Communications Interface) 303 and a communication bus 304, wherein, the processor 301, the memory 302, and the communication interface 303 complete mutual communication through the communication bus 304.
  • the processor 301 may invoke logic instructions in the memory 302 to perform the following method: acquire historical scheduling data in the historical records, and acquire data attributes and scheduling records corresponding to the historical data according to the historical scheduling data, the data attributes including Data source, data type, data size, and data purpose label.
  • the scheduling records include data migration, data deletion, and data merging; data containing data purpose labels in the historical data are excluded.
  • data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and transfer the data of the target data
  • the attribute is input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and data management is performed on the target data according to the scheduling record of the target data.
  • the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
  • the above-mentioned logic instructions in the memory 302 may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as an independent product.
  • the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical discs and other media that can store program codes.
  • an embodiment of the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored.
  • the transmission method provided by the above-mentioned embodiments is implemented, for example, including : Obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, the data attributes include data source, data type, data size, data purpose tag, the scheduling Records include data migration, data deletion, and data merging; exclude the data containing data purpose tags in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
  • the corresponding scheduling record is input into the convolutional neural network model as a training sample for training; the data attribute of the target data is obtained, and the data attribute of the target data is input into the trained convolutional neural network model, and the output corresponding to the target data
  • the scheduling record of the target data is used for data management of the target data according to the
  • each implementation can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware.
  • the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic CD, CD, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Library & Information Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Provided in the embodiments of the present invention are an artificial-intelligence-based data management method and apparatus for a data center. The method comprises: acquiring historical scheduling data from a historical record, and acquiring a data attribute and a scheduling record of corresponding historical data according to the historical scheduling data; excluding, from the historical data, data which includes a data destination label, and constructing a corresponding data attribute weight model according to a data source, the data type and the data size of the remaining historical data; taking the data attribute weight model and a corresponding scheduling record as training samples, and inputting same into a convolutional neural network model for training; and acquiring a data attribute of target data, inputting the data attribute of the target data into the trained convolutional neural network model, outputting a scheduling record corresponding to the target data, and performing data management on the target data according to the scheduling record of the target data. In this way, data management of target data can be completed according to the deep learning of artificial intelligence, thereby saving on management resources, and also improving the management efficiency.

Description

基于人工智能的数据中心数据管理方法及***Data center data management method and system based on artificial intelligence 技术领域technical field
本发明涉及人工智能技术领域，尤其涉及一种基于人工智能的数据中心数据管理方法及***。The present invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based data center data management method and system.
背景技术Background technique
目前，随着我国互联网技术的快速发展，数据网络平台的数据交互越来越多，进而导致处理数据的数据中心需要处理的数据量也越来越大，对数据中心的工作人员，以及数据中心的数据处理装置要求也越来越高。At present, with the rapid development of my country's Internet technology, there are more and more data interactions on the data network platform, which leads to an increasing amount of data to be processed by the data center, which affects the staff of the data center and the data center The requirements for data processing devices are also getting higher and higher.
技术问题technical problem
现有技术中，因为数据中心的数据管理越来越复杂，数据量越来越多，而需要对每个数据进行依次处理，数据中心要么通过多个数据中心进行配合处理，要么通过增加处理时间完成数据处理，但上述的处理手段不能在节约处理资源的同时，保证数据的处理效率。In the existing technology, because the data management of the data center is becoming more and more complex and the amount of data is increasing, each data needs to be processed sequentially. The data center either cooperates with multiple data centers or increases the processing time. Data processing is completed, but the above-mentioned processing means cannot guarantee the efficiency of data processing while saving processing resources.
技术解决方案technical solution
针对现有技术中存在的问题，本发明实施例提供一种基于人工智能的数据中心数据管理方法及***。Aiming at the problems existing in the prior art, embodiments of the present invention provide an artificial intelligence-based data center data management method and system.
本发明实施例提供一种基于人工智能的数据中心数据管理方法，包括：An embodiment of the present invention provides an artificial intelligence-based data center data management method, including:
获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；Obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, the data attributes include data source, data type, data size, data purpose label, and the scheduling records Including data migration, data deletion, data merging;
排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；Exclude the data containing the data purpose label in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；Input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training;
获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。Acquiring the data attributes of the target data, and inputting the data attributes of the target data into the trained convolutional neural network model, outputting the scheduling records corresponding to the target data, and performing data processing on the target data according to the scheduling records of the target data manage.
在其中一个实施例中，所述方法还包括：In one embodiment, the method also includes:
根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重；determining a first priority of the sender according to the data source, and determining a first weight of the remaining historical data according to the first priority;
根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重；determining a second priority of the remaining historical data according to the data type, and determining a second weight of the remaining historical data according to the second priority;
根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重；determining a third priority of the remaining historical data according to the data size, and determining a third weight of the remaining historical data according to the third priority;
根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A comprehensive weight is determined according to the first weight, the second weight, and the third weight, and a corresponding data attribute weight model is constructed according to the remaining historical data and the corresponding comprehensive weight.
在其中一个实施例中，所述方法还包括：In one embodiment, the method also includes:
获取数据中心当前容量，根据所述数据大小，结合所述当前容量，动态调整所述剩余历史数据的第三优先级。Obtain the current capacity of the data center, and dynamically adjust the third priority of the remaining historical data according to the size of the data and in combination with the current capacity.
在其中一个实施例中，所述方法还包括：In one embodiment, the method also includes:
获取预设的权重倾向，根据所述权重倾向，结合所述第一权重、第二权重、第三权重确定综合权重，所述权重倾向为发送方优先、优先级数据优先、数据中心性能优先三者之一。Obtain a preset weight tendency, and determine a comprehensive weight according to the weight tendency in combination with the first weight, the second weight, and the third weight, and the weight tendency is the sender priority, priority data priority, and data center performance priority three one of them.
在其中一个实施例中，所述方法还包括：In one embodiment, the method also includes:
将数据属性权重模型分别与所述数据迁移、数据删改、数据合并的调度记录作为训练样本输入到卷积神经网络模型中进行训练，分别得到三个训练卷积神经网络模型；Inputting the data attribute weight model and the scheduling record of the data migration, data deletion and data merging into the convolutional neural network model as training samples for training, respectively, to obtain three training convolutional neural network models;
所述将目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，包括：Said inputting the data attributes of the target data into the trained convolutional neural network model, and outputting the scheduling records corresponding to the target data, including:
将所述目标数据的数据属性分别输入至三个所述训练卷积神经网络模型，输出1个主要调度记录和2个额外记录。The data attributes of the target data are respectively input to the three training convolutional neural network models, and one main scheduling record and two additional records are output.
在其中一个实施例中，所述方法还包括：In one embodiment, the method also includes:
将所述训练样本分为训练集和验证集，将所述训练集输入到卷积神经网络模型进行训练，得到训练后的初步卷积神经网络模型；The training sample is divided into a training set and a verification set, and the training set is input to a convolutional neural network model for training to obtain a trained preliminary convolutional neural network model;
通过所述验证集输入到训练后的初步卷积神经网络模型进行测试，测试完成后得到所述训练后的卷积神经网络模型。The verification set is input to the trained preliminary convolutional neural network model for testing, and the trained convolutional neural network model is obtained after the test is completed.
本发明实施例提供一种基于人工智能的数据中心数据管理***，包括：An embodiment of the present invention provides an artificial intelligence-based data center data management system, including:
获取模块，用于获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；An acquisition module, configured to acquire historical scheduling data in historical records, and acquire data attributes and scheduling records of corresponding historical data according to the historical scheduling data, the data attributes including data sources, data types, data sizes, and data purpose tags , the scheduling record includes data migration, data deletion, and data merging;
排除模块，用于排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；An exclusion module, configured to exclude data containing data purpose tags in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
输入模块，用于将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；The input module is used to input the data attribute weight model and the corresponding scheduling record into the convolutional neural network model as training samples for training;
输出模块，用于获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。The output module is used to obtain the data attributes of the target data, and input the data attributes of the target data to the trained convolutional neural network model, output the scheduling records corresponding to the target data, and perform the scheduling according to the scheduling records of the target data. The above target data is used for data management.
在其中一个实施例中，所述***还包括：In one of the embodiments, the system also includes:
第一确定模块，用于根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重；A first determining module, configured to determine a first priority of the sender according to the data source, and determine a first weight of the remaining historical data according to the first priority;
第二确定模块，用于根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重；A second determining module, configured to determine a second priority of the remaining historical data according to the data type, and determine a second weight of the remaining historical data according to the second priority;
第三确定模块，用于根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重；A third determining module, configured to determine a third priority of the remaining historical data according to the data size, and determine a third weight of the remaining historical data according to the third priority;
构建模块，用于根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A construction module, configured to determine a comprehensive weight according to the first weight, the second weight, and the third weight, and construct a corresponding data attribute weight model according to the remaining historical data and the corresponding comprehensive weight.
本发明实施例提供一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现上述基于人工智能的数据中心数据管理方法的步骤。An embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor. When the processor executes the program, the above-mentioned artificial intelligence-based data center data management is realized. method steps.
本发明实施例提供一种非暂态计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现上述基于人工智能的数据中心数据管理方法的步骤。An embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned artificial intelligence-based data center data management method are implemented.
有益效果Beneficial effect
本发明实施例提供的基于人工智能的数据中心数据管理方法及***，获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。这样能够根据人工智能的深度学习完成对目标数据的数据管理，节省管理资源的同时，也提高了管理效率。The artificial intelligence-based data center data management method and system provided by the embodiments of the present invention acquire historical scheduling data in historical records, and acquire data attributes and scheduling records of corresponding historical data according to the historical scheduling data. The data attributes Including data sources, data types, data sizes, and data purpose tags, the scheduling records include data migration, data deletion, and data merging; data containing data purpose tags in the historical data are excluded, and according to the data sources and data of the remaining historical data Type and data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and The data attributes are input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and the target data is managed according to the scheduling record of the target data. In this way, the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
附图说明Description of drawings
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are For some embodiments of the present invention, those skilled in the art can also obtain other drawings based on these drawings without creative work.
图1为本发明实施例中基于人工智能的数据中心数据管理方法的流程图；Fig. 1 is the flow chart of the data center data management method based on artificial intelligence in the embodiment of the present invention;
图2为本发明实施例中基于人工智能的数据中心数据管理***的结构图；Fig. 2 is the structural diagram of the data center data management system based on artificial intelligence in the embodiment of the present invention;
图3为本发明实施例中电子设备结构示意图。FIG. 3 is a schematic structural diagram of an electronic device in an embodiment of the present invention.
本发明的实施方式Embodiments of the present invention
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.
图1为本发明实施例提供的基于人工智能的数据中心数据管理方法的流程示意图，如图1所示，本发明实施例提供了一种基于人工智能的数据中心数据管理方法，包括：Fig. 1 is a schematic flowchart of an artificial intelligence-based data center data management method provided by an embodiment of the present invention. As shown in Fig. 1 , the embodiment of the present invention provides an artificial intelligence-based data center data management method, including:
步骤S101，获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并。Step S101, obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, the data attributes include data source, data type, data size, data destination tag, all The above scheduling records include data migration, data deletion, and data merging.
具体地，获取数据中心中的历史记录，其中，获取历史记录中的历史调度数据，历史调度数据为历史数据本身的数据属性以及对应的数据调度记录，其中，数据属性可以包括数据来源、数据类型、数据大小、数据目的标签（数据发送目的地的标签），调度记录包括数据迁移、数据删改、数据合并。Specifically, the historical records in the data center are acquired, wherein the historical scheduling data in the historical records is acquired, and the historical scheduling data are the data attributes of the historical data itself and the corresponding data scheduling records, wherein the data attributes may include data sources, data types , data size, data destination label (the label of the data sending destination), and scheduling records include data migration, data deletion, and data merging.
步骤S102，排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型。Step S102, excluding data containing data purpose tags from the historical data, and constructing a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data.
具体地，确定历史数据中包含数据目的标签的数据后，即确定以及由明确数据处理目的地的数据后，对上述数据进行排除，然后对没有明确数据处理目的地的数据进行数据管理，通过剩余历史数据的数据来源、数据类型、数据大小为输入，构建对应的数据属性权重模型。Specifically, after it is determined that the historical data contains the data with the data purpose label, that is, after the data with a clear data processing destination is determined, the above data is excluded, and then data management is performed on the data without a clear data processing destination, through the remaining The data source, data type, and data size of historical data are input, and the corresponding data attribute weight model is constructed.
步骤S103，将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练。Step S103, input the data attribute weight model and the corresponding scheduling records as training samples into the convolutional neural network model for training.
具体地，将根据剩余历史数据构建的数据属性权重模型以及剩余历史数据对应的调度记录作为输入，输入到卷积神经网络模型的输入层进行模型训练，卷积神经网络模型通过卷积层-池化层-全连接层进行深度学习，得到训练后的卷积神经网络模型。Specifically, the data attribute weight model constructed according to the remaining historical data and the scheduling records corresponding to the remaining historical data are used as input, and input to the input layer of the convolutional neural network model for model training. The convolutional neural network model passes through the convolutional layer-pool Deep learning is performed on the layer-full connection layer to obtain the trained convolutional neural network model.
步骤S104，获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。Step S104: Obtain the data attribute of the target data, input the data attribute of the target data into the trained convolutional neural network model, output the scheduling record corresponding to the target data, and process the target according to the scheduling record of the target data Data for data management.
具体地，获取需要进行数据管理的目标数据的数据属性，并将目标数据的数据属性输入到训练后的卷积神经网络模型，并输出目标数据对应的调度记录，调度记录可以为对目标数据的数据迁移、数据删改、数据合并，根据数据的调度记录对目标数据进行数据管理。Specifically, the data attributes of the target data that need to be managed are obtained, and the data attributes of the target data are input into the trained convolutional neural network model, and the scheduling records corresponding to the target data are output. The scheduling records can be for the target data. Data migration, data deletion, data merging, and data management of the target data according to the data scheduling records.
另外，也可以将将数据属性权重模型分别与数据迁移、数据删改、数据合并的调度记录作为训练样本输入到卷积神经网络模型中进行训练，分别得到三个训练卷积神经网络模型，从而可以将目标数据的数据属性分别输入至三个训练卷积神经网络模型，输出1个主要调度记录和2个额外记录，其中，主要调度记录为三个记录中记录的管理方法更完整，更符合数据要求的记录，为主要数据管理方法，而2个额外记录记录的管理方法完整性要求不高，为目标数据的次要数据管理方法，进而提高数据管理的多样性。In addition, the scheduling records of the data attribute weight model and data migration, data deletion, and data merging can also be input into the convolutional neural network model as training samples for training, and three training convolutional neural network models are obtained respectively, so that Input the data attributes of the target data into three training convolutional neural network models respectively, and output 1 main scheduling record and 2 additional records. Among them, the management method of the main scheduling record being recorded in the three records is more complete and more in line with the data The required record is the main data management method, and the 2 additional records are the secondary data management method of the target data, which does not require high integrity of the management method of the record, thereby improving the diversity of data management.
本发明实施例提供的一种基于人工智能的数据中心数据管理方法，获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。这样能够根据人工智能的深度学习完成对目标数据的数据管理，节省管理资源的同时，也提高了管理效率。An artificial intelligence-based data center data management method provided by an embodiment of the present invention acquires historical scheduling data in historical records, and acquires data attributes and scheduling records of corresponding historical data according to the historical scheduling data. The data attributes Including data sources, data types, data sizes, and data purpose tags, the scheduling records include data migration, data deletion, and data merging; data containing data purpose tags in the historical data are excluded, and according to the data sources and data of the remaining historical data Type and data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and The data attributes are input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and the target data is managed according to the scheduling record of the target data. In this way, the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
在上述实施例的基础上，所述基于人工智能的数据中心数据管理方法，还包括：On the basis of the foregoing embodiments, the artificial intelligence-based data center data management method further includes:
根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重；determining a first priority of the sender according to the data source, and determining a first weight of the remaining historical data according to the first priority;
根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重；determining a second priority of the remaining historical data according to the data type, and determining a second weight of the remaining historical data according to the second priority;
根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重；determining a third priority of the remaining historical data according to the data size, and determining a third weight of the remaining historical data according to the third priority;
根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A comprehensive weight is determined according to the first weight, the second weight, and the third weight, and a corresponding data attribute weight model is constructed according to the remaining historical data and the corresponding comprehensive weight.
在本发明实施例中，在构建数据属性权重模型时，根据数据来源确定发送方的第一优先级，即根据发送方的优先级（比如客户的重要程度）确定第一优先级，进而确定历史数据的第一权重，然后根据数据类型确定历史数据的第二优先级，比如当数据类型的处理优先程度越高时，第二优先级也就越高，进而确定历史数据的第二权重，根据数据大小确定历史数据的第三优先级，第三优先级的确定可以通过获取数据中心当前容量，根据数据大小，结合当前容量，动态调整剩余历史数据的第三优先级，比如当数据中心当前容量出现预警（容量不足）时，历史数据的数据量越大，第三优先级也越高，进而确定历史数据的第三权重，在确定剩余历史数据的第一权重、第二权重、第三权重后，根据预设的权重倾向，结合第一权重、第二权重、第三权重确定综合权重，其中，权重倾向包括为发送方优先（即提高第一权重、降低第二权重、第三权重，比如第一权重、第二权重和第三权重的初始权重均为1/3，则根据权重倾向的程度提高提高第一权重、降低第二权重、第三权重）、优先级数据优先（即提高第二权重、降低第一权重、第三权重）、数据中心性能优先（即提高第三权重、降低第一权重、第二权重）三者之一， 并根据剩余历史数据中数据的数据来源、数据类型、数据大小与综合权重质检的对应关系建立对应的数据属性权重模型。In the embodiment of the present invention, when constructing the data attribute weight model, the first priority of the sender is determined according to the source of the data, that is, the first priority is determined according to the priority of the sender (such as the importance of the customer), and then the history The first weight of the data, and then determine the second priority of the historical data according to the data type, for example, the higher the processing priority of the data type, the higher the second priority, and then determine the second weight of the historical data, according to The data size determines the third priority of historical data. The third priority can be determined by obtaining the current capacity of the data center, according to the size of the data, combined with the current capacity, and dynamically adjusting the third priority of the remaining historical data. For example, when the current capacity of the data center When an early warning (insufficient capacity) occurs, the larger the amount of historical data, the higher the third priority, and then determine the third weight of historical data. After determining the first weight, second weight, and third weight of the remaining historical data Finally, according to the preset weight tendency, combined with the first weight, the second weight, and the third weight to determine the comprehensive weight, wherein, the weight tendency includes giving priority to the sender (that is, increasing the first weight, reducing the second weight, and the third weight, For example, the initial weights of the first weight, the second weight, and the third weight are all 1/3, and the first weight is increased, the second weight is decreased, and the third weight is increased according to the degree of weight tendency, and the priority data is given priority (that is, increase second weight, reduce the first weight, third weight), data center performance priority (that is, increase the third weight, reduce the first weight, second weight) one of the three, and according to the data source of the data in the remaining historical data, The corresponding relationship between data type, data size and comprehensive weight quality inspection establishes a corresponding data attribute weight model.
本发明实施例通过剩余历史数据的三项权重构建对应的数据属性权重模型，进而通过数据属性权重模型完成人工智能的深度学习管理。In the embodiment of the present invention, the corresponding data attribute weight model is constructed through the three weights of the remaining historical data, and then the deep learning management of artificial intelligence is completed through the data attribute weight model.
在上述实施例的基础上，所述基于人工智能的数据中心数据管理方法，还包括：On the basis of the foregoing embodiments, the artificial intelligence-based data center data management method further includes:
将所述训练样本分为训练集和验证集，将所述训练集输入到卷积神经网络模型进行训练，得到训练后的初步卷积神经网络模型；The training sample is divided into a training set and a verification set, and the training set is input to a convolutional neural network model for training to obtain a trained preliminary convolutional neural network model;
通过所述验证集输入到训练后的初步卷积神经网络模型进行测试，测试完成后得到所述训练后的卷积神经网络模型。The verification set is input to the trained preliminary convolutional neural network model for testing, and the trained convolutional neural network model is obtained after the test is completed.
在本发明实施例中，在对卷积神经网络模型进行训练时，通常会分出一部分样本数据，作为验证集，对训练后的卷积神经网络模型的数据准确性进行验证，当对数据属性权重模型与对应的调度记录进行训练时，对数据属性权重模型与对应的调度记录的训练样本进行数据分组，具体可以将样本数据分为80%的训练集与20%的验证集，通过训练集与训练集数据对应的数据属性权重模型和调度记录进行初步训练，得到初步卷积神经网络模型，然后通过验证集与验证集数据对应的数据属性权重模型和调度记录对初步卷积神经网络模型进行测试，得到训练后的卷积神经网络模型。In the embodiment of the present invention, when training the convolutional neural network model, a part of sample data is usually separated as a verification set to verify the data accuracy of the trained convolutional neural network model. When the data attribute When the weight model and the corresponding scheduling records are trained, the training samples of the data attribute weight model and the corresponding scheduling records are grouped. Specifically, the sample data can be divided into 80% of the training set and 20% of the verification set. Through the training set Preliminary training is performed on the data attribute weight model and scheduling records corresponding to the training set data to obtain the preliminary convolutional neural network model, and then the preliminary convolutional neural network model is trained through the verification set and the data attribute weight model and scheduling records corresponding to the verification set data. Test to get the trained convolutional neural network model.
本发明实施例通过将样本数据进行数据分组，通过训练集建立初步模型，通过验证集对初步模型进行准确性验证，保证了卷积神经网络模型的准确性。In the embodiment of the present invention, the accuracy of the convolutional neural network model is ensured by grouping the sample data into data groups, establishing a preliminary model through the training set, and verifying the accuracy of the preliminary model through the verification set.
图2为本发明实施例提供的一种基于人工智能的计量管理***，包括：获取模块S201、排除模块S202、输入模块S203、输出模块S204，其中：Fig. 2 is an artificial intelligence-based metering management system provided by an embodiment of the present invention, including: an acquisition module S201, an exclusion module S202, an input module S203, and an output module S204, wherein:
获取模块S201，用于获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并。The obtaining module S201 is used to obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, and the data attributes include data source, data type, data size, and data purpose label, the scheduling records include data migration, data deletion, and data merging.
排除模块S202，用于排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型。The exclusion module S202 is configured to exclude data containing data purpose tags from the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data.
输入模块S203，用于将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练。The input module S203 is used for inputting the data attribute weight model and the corresponding scheduling records as training samples into the convolutional neural network model for training.
输出模块S204，用于获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。The output module S204 is used to obtain the data attributes of the target data, and input the data attributes of the target data into the trained convolutional neural network model, and output the scheduling records corresponding to the target data, according to the scheduling records of the target data The target data is subject to data management.
在一个实施例中，***还可以包括：In one embodiment, the system may also include:
第一确定模块，用于根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重。A first determining module, configured to determine a first priority of the sender according to the data source, and determine a first weight of the remaining historical data according to the first priority.
第二确定模块，用于根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重。The second determining module is configured to determine a second priority of the remaining historical data according to the data type, and determine a second weight of the remaining historical data according to the second priority.
第三确定模块，用于根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重。A third determining module, configured to determine a third priority of the remaining historical data according to the data size, and determine a third weight of the remaining historical data according to the third priority.
构建模块，用于根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A construction module, configured to determine a comprehensive weight according to the first weight, the second weight, and the third weight, and construct a corresponding data attribute weight model according to the remaining historical data and the corresponding comprehensive weight.
在一个实施例中，***还可以包括：In one embodiment, the system may also include:
第二获取模块，用于获取数据中心当前容量，根据所述数据大小，结合所述当前容量，动态调整所述剩余历史数据的第三优先级。The second acquisition module is configured to acquire the current capacity of the data center, and dynamically adjust the third priority of the remaining historical data according to the size of the data and in combination with the current capacity.
在一个实施例中，***还可以包括：In one embodiment, the system may also include:
第三获取模块，用于获取预设的权重倾向，根据所述权重倾向，结合所述第一权重、第二权重、第三权重确定综合权重，所述权重倾向为发送方优先、优先级数据优先、数据中心性能优先三者之一。The third acquisition module is used to obtain a preset weight tendency, and according to the weight tendency, combined with the first weight, the second weight, and the third weight to determine a comprehensive weight, the weight tendency is sender priority, priority data One of priority and data center performance priority.
在一个实施例中，***还可以包括：In one embodiment, the system may also include:
训练模块，用于将所述训练样本分为训练集和验证集，将所述训练集输入到卷积神经网络模型进行训练，得到训练后的初步卷积神经网络模型。The training module is used to divide the training samples into a training set and a verification set, input the training set into the convolutional neural network model for training, and obtain a trained preliminary convolutional neural network model.
测试模块，用于通过所述验证集输入到训练后的初步卷积神经网络模型进行测试，测试完成后得到所述训练后的卷积神经网络模型。The test module is used for inputting the verification set into the trained preliminary convolutional neural network model for testing, and obtaining the trained convolutional neural network model after the test is completed.
关于基于人工智能的数据中心数据管理***的具体限定可以参见上文中对于基于人工智能的数据中心数据管理方法的限定，在此不再赘述。上述基于人工智能的数据中心数据管理***中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。For the specific limitations of the artificial intelligence-based data center data management system, please refer to the above-mentioned definition of the artificial intelligence-based data center data management method, which will not be repeated here. Each module in the above artificial intelligence-based data center data management system can be realized in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.
图3示例了一种电子设备的实体结构示意图，如图3所示，该电子设备可以包括：处理器(processor)301、存储器(memory)302、通信接口(Communications Interface)303和通信总线304，其中，处理器301，存储器302，通信接口303通过通信总线304完成相互间的通信。处理器301可以调用存储器302中的逻辑指令，以执行如下方法：获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。这样能够根据人工智能的深度学习完成对目标数据的数据管理，节省管理资源的同时，也提高了管理效率。 FIG. 3 illustrates a schematic diagram of the physical structure of an electronic device. As shown in FIG. 3, the electronic device may include: a processor (processor) 301, a memory (memory) 302, a communication interface (Communications Interface) 303 and a communication bus 304, wherein, the processor 301, the memory 302, and the communication interface 303 complete mutual communication through the communication bus 304. The processor 301 may invoke logic instructions in the memory 302 to perform the following method: acquire historical scheduling data in the historical records, and acquire data attributes and scheduling records corresponding to the historical data according to the historical scheduling data, the data attributes including Data source, data type, data size, and data purpose label. The scheduling records include data migration, data deletion, and data merging; data containing data purpose labels in the historical data are excluded. According to the data source and data type of the remaining historical data, , data size to construct a corresponding data attribute weight model; input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training; obtain the data attribute of the target data, and transfer the data of the target data The attribute is input to the trained convolutional neural network model, and the scheduling record corresponding to the target data is output, and data management is performed on the target data according to the scheduling record of the target data. In this way, the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency. the
此外，上述的存储器302中的逻辑指令可以通过软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备（可以是个人计算机，服务器，或者网络设备等）执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器（ROM，Read-Only Memory）、随机存取存储器（RAM，Random Access Memory）、磁碟或者光盘等各种可以存储程序代码的介质。In addition, the above-mentioned logic instructions in the memory 302 may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical discs and other media that can store program codes.
另一方面，本发明实施例还提供一种非暂态计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现以执行上述各实施例提供的传输方法，例如包括：获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。这样能够根据人工智能的深度学习完成对目标数据的数据管理，节省管理资源的同时，也提高了管理效率。On the other hand, an embodiment of the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the transmission method provided by the above-mentioned embodiments is implemented, for example, including : Obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, the data attributes include data source, data type, data size, data purpose tag, the scheduling Records include data migration, data deletion, and data merging; exclude the data containing data purpose tags in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data; The corresponding scheduling record is input into the convolutional neural network model as a training sample for training; the data attribute of the target data is obtained, and the data attribute of the target data is input into the trained convolutional neural network model, and the output corresponding to the target data The scheduling record of the target data is used for data management of the target data according to the scheduling record of the target data. In this way, the data management of the target data can be completed according to the deep learning of artificial intelligence, which saves management resources and improves management efficiency.
以上所描述的***实施例仅仅是示意性的，其中所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。The system embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备（可以是个人计算机，服务器，或者网络设备等）执行各个实施例或者实施例的某些部分所述的方法。Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic CD, CD, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (10)

  1. 一种基于人工智能的数据中心数据管理方法，其特征在于，包括：An artificial intelligence-based data center data management method is characterized in that it includes:
    获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；Obtain the historical scheduling data in the historical records, and obtain the data attributes and scheduling records of the corresponding historical data according to the historical scheduling data, the data attributes include data source, data type, data size, data purpose label, and the scheduling records Including data migration, data deletion, data merging;
    排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；Exclude the data containing the data purpose label in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
    将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；Input the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training;
    获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。Acquiring the data attributes of the target data, and inputting the data attributes of the target data into the trained convolutional neural network model, outputting the scheduling records corresponding to the target data, and performing data processing on the target data according to the scheduling records of the target data manage.
  2. 根据权利要求1所述的基于人工智能的数据中心数据管理方法，其特征在于，所述根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型，包括：The artificial intelligence-based data center data management method according to claim 1, wherein said constructing a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data includes:
    根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重；determining a first priority of the sender according to the data source, and determining a first weight of the remaining historical data according to the first priority;
    根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重；determining a second priority of the remaining historical data according to the data type, and determining a second weight of the remaining historical data according to the second priority;
    根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重；determining a third priority of the remaining historical data according to the data size, and determining a third weight of the remaining historical data according to the third priority;
    根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A comprehensive weight is determined according to the first weight, the second weight, and the third weight, and a corresponding data attribute weight model is constructed according to the remaining historical data and the corresponding comprehensive weight.
  3. 根据权利要求2所述的基于人工智能的数据中心数据管理方法，其特征在于，所述根据数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重，包括：The artificial intelligence-based data center data management method according to claim 2, wherein the third priority of the remaining historical data is determined according to the data size, and the remaining historical data is determined according to the third priority The third weight of the data, including:
    获取数据中心当前容量，根据所述数据大小，结合所述当前容量，动态调整所述剩余历史数据的第三优先级。Obtain the current capacity of the data center, and dynamically adjust the third priority of the remaining historical data according to the size of the data and in combination with the current capacity.
  4. 根据权利要求2所述的基于人工智能的数据中心数据管理方法，其特征在于，所述根据第一权重、第二权重、第三权重确定综合权重，包括：The artificial intelligence-based data center data management method according to claim 2, wherein said determining the comprehensive weight according to the first weight, the second weight, and the third weight includes:
    获取预设的权重倾向，根据所述权重倾向，结合所述第一权重、第二权重、第三权重确定综合权重，所述权重倾向为发送方优先、优先级数据优先、数据中心性能优先三者之一。Obtain a preset weight tendency, and determine a comprehensive weight according to the weight tendency in combination with the first weight, the second weight, and the third weight, and the weight tendency is the sender priority, priority data priority, and data center performance priority three one of them.
  5. 根据权利要求1所述的基于人工智能的数据中心数据管理方法，其特征在于，将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练，包括：The artificial intelligence-based data center data management method according to claim 1, wherein the data attribute weight model and corresponding scheduling records are input into the convolutional neural network model as training samples for training, including:
    将数据属性权重模型分别与所述数据迁移、数据删改、数据合并的调度记录作为训练样本输入到卷积神经网络模型中进行训练，分别得到三个训练卷积神经网络模型；Inputting the data attribute weight model and the scheduling record of the data migration, data deletion and data merging into the convolutional neural network model as training samples for training, respectively, to obtain three training convolutional neural network models;
    所述将目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，包括：Said inputting the data attributes of the target data into the trained convolutional neural network model, and outputting the scheduling records corresponding to the target data, including:
    将所述目标数据的数据属性分别输入至三个所述训练卷积神经网络模型，输出1个主要调度记录和2个额外记录。The data attributes of the target data are respectively input to the three training convolutional neural network models, and one main scheduling record and two additional records are output.
  6. 根据权利要求1所述的基于人工智能的数据中心数据管理方法，其特征在于，所述将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练，包括：The artificial intelligence-based data center data management method according to claim 1, wherein said inputting the data attribute weight model and corresponding scheduling records as training samples into the convolutional neural network model for training includes:
    将所述训练样本分为训练集和验证集，将所述训练集输入到卷积神经网络模型进行训练，得到训练后的初步卷积神经网络模型；The training sample is divided into a training set and a verification set, and the training set is input to a convolutional neural network model for training to obtain a trained preliminary convolutional neural network model;
    通过所述验证集输入到训练后的初步卷积神经网络模型进行测试，测试完成后得到所述训练后的卷积神经网络模型。The verification set is input to the trained preliminary convolutional neural network model for testing, and the trained convolutional neural network model is obtained after the test is completed.
  7. 一种基于人工智能的数据中心数据管理***，其特征在于，所述***包括：A data center data management system based on artificial intelligence, characterized in that the system includes:
    获取模块，用于获取历史记录中的历史调度数据，并根据所述历史调度数据获取对应的历史数据的数据属性和调度记录，所述数据属性包括数据来源、数据类型、数据大小、数据目的标签，所述调度记录包括数据迁移、数据删改、数据合并；An acquisition module, configured to acquire historical scheduling data in historical records, and acquire data attributes and scheduling records of corresponding historical data according to the historical scheduling data, the data attributes including data sources, data types, data sizes, and data purpose tags , the scheduling record includes data migration, data deletion, and data merging;
    排除模块，用于排除所述历史数据中包含数据目的标签的数据，根据剩余历史数据的数据来源、数据类型、数据大小构建对应的数据属性权重模型；An exclusion module, configured to exclude data containing data purpose tags in the historical data, and construct a corresponding data attribute weight model according to the data source, data type, and data size of the remaining historical data;
    输入模块，用于将数据属性权重模型与对应的调度记录作为训练样本输入到卷积神经网络模型中进行训练；The input module is used to input the data attribute weight model and the corresponding scheduling record into the convolutional neural network model as training samples for training;
    输出模块，用于获取目标数据的数据属性，并将所述目标数据的数据属性输入至训练后的卷积神经网络模型，输出目标数据对应的调度记录，根据所述目标数据的调度记录对所述目标数据进行数据管理。The output module is used to obtain the data attributes of the target data, and input the data attributes of the target data to the trained convolutional neural network model, output the scheduling records corresponding to the target data, and perform the scheduling according to the scheduling records of the target data. The above target data is used for data management.
  8. 根据权利要求7中所述的基于人工智能的数据中心数据管理装置，其特征在于，所述装置还包括：According to the artificial intelligence-based data center data management device described in claim 7, it is characterized in that, the device also includes:
    第一确定模块，用于根据所述数据来源确定发送方的第一优先级，根据所述第一优先级确定所述剩余历史数据的第一权重；A first determining module, configured to determine a first priority of the sender according to the data source, and determine a first weight of the remaining historical data according to the first priority;
    第二确定模块，用于根据所述数据类型确定所述剩余历史数据的第二优先级，根据所述第二优先级确定所述剩余历史数据的第二权重；A second determining module, configured to determine a second priority of the remaining historical data according to the data type, and determine a second weight of the remaining historical data according to the second priority;
    第三确定模块，用于根据所述数据大小确定所述剩余历史数据的第三优先级，根据所述第三优先级确定所述剩余历史数据的第三权重；A third determining module, configured to determine a third priority of the remaining historical data according to the data size, and determine a third weight of the remaining historical data according to the third priority;
    构建模块，用于根据所述第一权重、第二权重、第三权重确定综合权重，根据所述剩余历史数据及对应的综合权重构建对应的数据属性权重模型。A construction module, configured to determine a comprehensive weight according to the first weight, the second weight, and the third weight, and construct a corresponding data attribute weight model according to the remaining historical data and the corresponding comprehensive weight.
  9. 一种电子设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时实现如权利要求1至6任一项所述基于人工智能的数据中心数据管理方法的步骤。An electronic device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, characterized in that, when the processor executes the program, the computer program described in any one of claims 1 to 6 is implemented. Describe the steps of the artificial intelligence-based data center data management method.
  10. 一种非暂态计算机可读存储介质，其上存储有计算机程序，其特征在于，该计算机程序被处理器执行时实现如权利要求1至6任一项所述基于人工智能的数据中心数据管理方法的步骤。A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the artificial intelligence-based data center data management according to any one of claims 1 to 6 is realized method steps.
PCT/CN2021/103347 2021-06-25 2021-06-30 Artificial-intelligence-based data management method and system for data center WO2022267085A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110708644.6 2021-06-25
CN202110708644.6A CN113536081B (en) 2021-06-25 2021-06-25 Data center data management method and system based on artificial intelligence

Publications (1)

Publication Number Publication Date
WO2022267085A1 true WO2022267085A1 (en) 2022-12-29

Family

ID=78096702

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/103347 WO2022267085A1 (en) 2021-06-25 2021-06-30 Artificial-intelligence-based data management method and system for data center

Country Status (2)

Country Link
CN (1) CN113536081B (en)
WO (1) WO2022267085A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116739317A (en) * 2023-08-15 2023-09-12 山东宇飞传动技术有限公司 Mining winch automatic management and dispatching platform, method, equipment and medium
CN116993307A (en) * 2023-09-28 2023-11-03 广东省信息工程有限公司 Collaborative office method and system with artificial intelligence learning capability
CN117235731A (en) * 2023-09-04 2023-12-15 广东立升科技有限公司 Big data monitoring and early warning system for secret equipment

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117527928B (en) * 2023-11-07 2024-05-10 河北燃气有限公司 SCADA-based data processing method and system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140343711A1 (en) * 2013-05-17 2014-11-20 The Research Foundation Of State University Of New York Decision support system for order prioritization
CN109146707A (en) * 2018-08-27 2019-01-04 罗孚电气（厦门）有限公司 Power consumer analysis method, device and electronic equipment based on big data analysis
CN109816412A (en) * 2017-11-21 2019-05-28 腾讯科技（深圳）有限公司 A kind of training pattern generation method, device, equipment and computer storage medium

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106874219B (en) * 2016-12-23 2018-11-02 深圳云天励飞技术有限公司 A kind of data dispatching method of convolutional neural networks, system and computer equipment
TWI650714B (en) * 2018-01-19 2019-02-11 財團法人工業技術研究院 Dynamic intelligent scheduling method and device
CN110363621B (en) * 2019-06-28 2022-04-12 东莞盟大集团有限公司 Order information pushing system based on artificial intelligence technology

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140343711A1 (en) * 2013-05-17 2014-11-20 The Research Foundation Of State University Of New York Decision support system for order prioritization
CN109816412A (en) * 2017-11-21 2019-05-28 腾讯科技（深圳）有限公司 A kind of training pattern generation method, device, equipment and computer storage medium
CN109146707A (en) * 2018-08-27 2019-01-04 罗孚电气（厦门）有限公司 Power consumer analysis method, device and electronic equipment based on big data analysis

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116739317A (en) * 2023-08-15 2023-09-12 山东宇飞传动技术有限公司 Mining winch automatic management and dispatching platform, method, equipment and medium
CN116739317B (en) * 2023-08-15 2023-10-31 山东宇飞传动技术有限公司 Mining winch automatic management and dispatching platform, method, equipment and medium
CN117235731A (en) * 2023-09-04 2023-12-15 广东立升科技有限公司 Big data monitoring and early warning system for secret equipment
CN117235731B (en) * 2023-09-04 2024-02-09 广东立升科技有限公司 Big data monitoring and early warning system for secret equipment
CN116993307A (en) * 2023-09-28 2023-11-03 广东省信息工程有限公司 Collaborative office method and system with artificial intelligence learning capability
CN116993307B (en) * 2023-09-28 2024-01-05 广东省信息工程有限公司 Collaborative office method and system with artificial intelligence learning capability

Also Published As

Publication number Publication date
CN113536081A (en) 2021-10-22
CN113536081B (en) 2022-04-01

Similar Documents

Publication Publication Date Title
WO2022267085A1 (en) Artificial-intelligence-based data management method and system for data center
WO2021115480A1 (en) Federated learning method, device, equipment, and storage medium
US20190378016A1 (en) Distributed computing architecture for large model deep learning
TWI735545B (en) Model training method and device
US11551437B2 (en) Collaborative information extraction
CN111061881A (en) Text classification method, equipment and storage medium
US11388244B1 (en) Method, device, and program product for managing computer system
JP2023534696A (en) Anomaly detection in network topology
WO2024037368A1 (en) Scheduling optimization method of scheduling apparatus, scheduling apparatus and storage medium
CN111611622A (en) Block chain-based file storage method and electronic equipment
US20220012642A1 (en) Dynamic network bandwidth in distributed deep learning training
US10320698B1 (en) Determining network connectivity for placement decisions
TWI758223B (en) Computing method with dynamic minibatch sizes and computing system and computer-readable storage media for performing the same
US11556827B2 (en) Transferring large datasets by using data generalization
CN111444148B (en) Data transmission method and device based on MapReduce
WO2024027071A1 (en) Data monitoring method and system
WO2023174189A1 (en) Method and apparatus for classifying nodes of graph network model, and device and storage medium
CN111445027B (en) Training method and device for machine learning model
CN108711074B (en) Service classification method, device, server and readable storage medium
WO2023278043A1 (en) Method and system for resource governance in a multi-tenant system
JP2023002475A (en) Computer system, computer program and computer-implemented method (causal knowledge identification and extraction)
US10986230B1 (en) Method and apparatus to capture, analyze, organize, and present support chat information
CN113760680A (en) Method and device for testing system pressure performance
TWI574169B (en) Method of operating a solution searching system and solution searching system
CN116915870B (en) Task creation request processing method, device, electronic equipment and readable medium

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21946574

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE