Data transmission early warning method and device

Abstract

The application discloses a data transmission early warning method and device. The method comprises the following steps: a central server trains an initial model by using a first historical network parameter pre-stored in the central server to obtain a first prediction model; the central server sends the first prediction model to a plurality of remote servers; the central server controls the plurality of remote servers to train the received first prediction model by using a second historical network parameter stored in the remote servers to obtain a plurality of second prediction models; and the central server receives the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are used for monitoring network parameters in a data transmission process of the plurality of remote servers, and early warning is performed when a monitoring result meets a preset condition. The application solves the technical problem that fault early warning in the data transmission process is not timely.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a data transmission early warning method and device. BACKGROUND

[0002] With the rapid development of computer technology, information networks have become an important guarantee for social development. The continuous improvement of network social development and the increasing popularity of network applications. Large-scale application of artificial intelligence technology, virtualization of traditional hardware resources, and centralized management and operation and maintenance system of virtual resources, business resources and user resources, virtual data is more complex, which brings great difficulty to the management of data transmission, and abnormal conditions in the data transmission process cannot be found in time.

[0003] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The embodiments of the present application provide a data transmission early warning method and device to at least solve the technical problem of not timely fault warning in the data transmission process.

[0005] According to an aspect of an embodiment of the present application, a data transmission early warning method is provided, comprising: a central server training an initial model by using a first historical network parameter pre-stored in the central server to obtain a first prediction model; the central server sending the first prediction model to a plurality of remote servers; the central server controlling the plurality of remote servers to train the received first prediction model by using a second historical network parameter stored in the remote server to obtain a plurality of second prediction models; and the central server receiving the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are used to monitor network parameters in the data transmission process of the plurality of remote servers, and perform early warning if the monitoring result meets a preset condition.

[0006] Optionally, the central server trains the first prediction model by using the first historical network parameter pre-stored in the central server to obtain a central prediction model, comprising: obtaining the first historical network parameter, wherein the first historical network parameter at least includes: resource occupancy rate at each sampling time in a first preset period and network delay at each sampling time in the first preset period; and using network parameters at a plurality of times in a second preset period before the time of data transmission anomaly in the first preset period as labels, and using network parameters at each sampling time in the first preset period as training data set to train the first prediction model.

[0007] Optionally, the obtaining the first historical network parameter comprises: obtaining the first historical network parameter from a database in the central server; and in a case where one or more time point data are missing in the first historical network parameter, determining the missing data values based on data of adjacent multiple time points before and after the missing data by using a second difference method.

[0008] Optionally, after the first historical network parameter is obtained, the method further comprises: performing feature extraction on spatial information and time information of the first historical network parameter respectively to obtain spatial features and time features; weighting the spatial features and the time features by using an attention mechanism to obtain fused features; and inputting the fused features into an initial model to train the initial model to obtain the first prediction model.

[0009] Optionally, before the central server controls the multiple remote servers to train the received first prediction model by using the second historical network parameter stored in the remote servers to obtain multiple second prediction models, the method further comprises: the central server sending a confirmation instruction to the multiple remote servers to determine whether the first prediction model exists in the multiple remote servers; and in a case where the first prediction model is not received in the remote servers, the central server sends the first prediction model to the remote servers again which do not receive the first prediction model.

[0010] Optionally, the multiple second prediction models are used for monitoring network parameters in a data transmission process of the multiple remote servers, and in a case where a monitoring result meets a preset condition, a warning is performed, comprising: the central server controlling the multiple remote servers to obtain network parameters in the data transmission process of the multiple remote servers; and the central server controlling the remote servers to transmit the network parameters to the remote prediction models, and when the remote prediction models detect that there are unverified parameters in the network parameters, a warning information is sent.

[0011] According to another aspect of the embodiments of the present application, another data transmission warning method is also provided, comprising: a remote server receiving a first prediction model sent by a central server; the remote server training the received first prediction model by using a second historical network parameter stored in the remote server to obtain a second prediction model; and the remote server sending the trained second prediction model to the central server, wherein the second prediction model is used for monitoring network parameters in a data transmission process of the remote server, and in a case where a monitoring result meets a preset condition, a warning is performed.

[0012] According to a further aspect of the embodiments of the present application, a data transmission early warning device is also provided, comprising: a first training module configured to train an initial model by using first historical network parameters pre-stored in a central server to obtain a first prediction model; a sending module configured to send the first prediction model to a plurality of remote servers; a second training module configured to control the plurality of remote servers to train the received first prediction model by using second historical network parameters stored in the remote servers to obtain a plurality of second prediction models; and an early warning module configured to receive the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are configured to monitor network parameters in a data transmission process of the plurality of remote servers, and perform early warning when a monitoring result meets a preset condition.

[0013] According to a further aspect of the embodiments of the present application, a non-volatile storage medium is also provided, comprising a stored program, wherein the program, when executed, controls a device in which the non-volatile storage medium is located to perform the data transmission early warning method.

[0014] According to a further aspect of the embodiments of the present application, an electronic device is also provided, comprising a memory and a processor, wherein the processor is configured to execute a program, and the program, when executed, performs the data transmission early warning method.

[0015] In the embodiments of the present application, the central server trains an initial model by using first historical network parameters pre-stored in the central server to obtain a first prediction model, sends the first prediction model to a plurality of remote servers, controls the plurality of remote servers to train the received first prediction model by using second historical network parameters stored in the remote servers to obtain a plurality of second prediction models, and receives the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are configured to monitor network parameters in a data transmission process of the plurality of remote servers, and perform early warning when a monitoring result meets a preset condition. In this way, by training an initial model by using first historical network parameters stored in the central server and then training by using second historical network parameters stored in the plurality of remote servers, a second prediction model suitable for the situation of each remote server is obtained, and then the second prediction model is used to monitor the data transmission process in real time, so that the purpose of locating a fault before the data transmission fails is achieved, thereby achieving the technical effect of improving the early warning efficiency, and further solving the technical problem of untimely fault early warning in the data transmission process. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of a computer terminal (or mobile device) for a data transmission early warning method according to an embodiment of the application;

[0018] Figure 2 is a flow chart of a data transmission early warning method according to the application;

[0019] Figure 3 is another optional data transmission early warning method flow chart according to an embodiment of the application;

[0020] Figure 4 is an optional data transmission early warning device schematic diagram according to the application. DETAILED DESCRIPTION

[0021] In order to make the personnel in the technical field better understand the application scheme, the technical scheme in the embodiments of the application will be clearly and completely described below in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor should be within the scope of protection of the application.

[0022] It should be noted that the terms "first", "second", and the like in the specification and claims of the application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to an embodiment of the application, an embodiment of a data transmission early warning method is also provided. It should be noted that the steps shown in the flow chart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flow chart, in some cases, the steps shown or described herein can be executed in a different order than that shown herein.

[0024] The method embodiments provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal, a cloud server or similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing the data transmission early warning method is shown. As shown in Figure 1 The computer terminal 10 (or mobile device 10) can include one or more processors 102 (the processor 102 can include but is not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the I / O interface), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can include more or fewer components than those shown in Figure 1 or have a different configuration than that shown in Figure 1 .

[0025] It should be noted that the one or more processors 102 and / or other data processing circuits described above can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or any one of the other elements combined into the computer terminal 10 (or mobile device) in whole or in part. As referred to in the embodiments of the present application, the data processing circuit serves as a processor control (for example, selection of a variable resistance terminal path connected to an interface).

[0026] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the data transmission early warning method in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the above-mentioned data transmission early warning method. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory such as one or more magnetic storage devices, flash memory or other non-volatile solid state memory. In some examples, the memory 104 can further include a memory remotely disposed with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0027] The transmission module 106 is configured to receive or send data via a network. The network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission module 106 can be a radio frequency (RF) module that is configured to communicate with the Internet wirelessly.

[0028] The display can be a liquid crystal display (LCD) that is touch screen, for example, which can enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0029] According to an embodiment of the present application, an embodiment of a data transmission early warning method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0030] Figure 2 is a flowchart of a data transmission early warning method according to an embodiment of the present application, as shown in Figure 2 The method comprises the following steps:

[0031] In step S202, the central server trains an initial model using a first historical network parameter pre-stored in the central server to obtain a first prediction model.

[0032] In step S204, the central server sends the first prediction model to a plurality of remote servers.

[0033] In step S206, the central server controls the plurality of remote servers to train the received first prediction model using a second historical network parameter stored in the remote servers to obtain a plurality of second prediction models.

[0034] In step S208, the central server receives the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are used to monitor network parameters in a data transmission process of the plurality of remote servers, and perform early warning when a monitoring result meets a preset condition.

[0035] Through the above steps, the second prediction model suitable for the situation of each remote server itself can be obtained by training the initial model by using the first historical network parameters stored in the central server and then training by using the second historical network parameters stored in the plurality of remote servers, and then the process of data transmission is monitored in real time by using the second prediction model, so that the purpose of locating the fault before the data transmission fails is achieved, thereby realizing the technical effect of improving the early warning efficiency, and further solving the technical problem of not timely fault early warning in the data transmission process.

[0036] It should be noted that the early warning method provided by the present application also reduces the computing pressure of the central server by distributing the training task to the plurality of remote servers, and at the same time, since the second prediction model is trained by the remote server by using the historical network parameters stored in the remote server, the security of the historical network parameters stored in the remote server is improved.

[0037] It should be further noted that the improved method of the present application can be applied in the scene of multi-cloud management to manage a plurality of cloud data.

[0038] The initial model in step S202 can be constructed by using a Transformer (The problem of sequence transduction) model.

[0039] In step S206, the preset condition includes but is not limited to: monitoring that the CPU occupancy rate is higher than a preset threshold, the network delay is higher than a set threshold, etc.

[0040] The above steps S202 to S208 will be described in detail by specific embodiments.

[0041] In step S202, the central server trains the first prediction model by using the first historical network parameters pre-stored in the central server to obtain a central prediction model, including: obtaining the first historical network parameters, wherein the first historical network parameters at least include: resource occupancy rate at each sampling time in a first preset period and network delay at each sampling time in the first preset period; using the network parameters of a plurality of time points in a second preset period before the time point of data transmission anomaly in the first preset period as labels, and using the network parameters at each sampling time in the first preset period as a training data set to train the first prediction model.

[0042] By using the network parameters of a plurality of time points in a second preset period before the time point of data transmission anomaly in the first preset period as labels, the condition of the network parameters in the second preset period before the transmission anomaly is determined, and then whether the data transmission will fail is determined by the network parameters.

[0043] It needs to be explained that the manifestation of the data transmission exception includes but is not limited to data transmission interruption, data transmission rate lower than the preset rate threshold, etc., and the resource occupation rate includes but is not limited to: CPU occupation rate; the length of the first preset period is set based on the amount of the training data set, and the length of the second preset period can be determined according to the early warning length.

[0044] In an optional mode, the first historical network parameter can be obtained from the database in the central server, and in the case that there is one or more missing data at one or more time points in the first historical network parameter, the missing data value is determined based on the data of the adjacent multiple time points before and after the missing data by using the second difference method.

[0045] Taking the determination of the missing data value from the data of the adjacent three time points before and after the missing data as an example, the missing data value is determined according to the following formula:

[0046]

[0047] In the formula, y represents the missing data value, x represents the network parameter corresponding to the time point of the missing data value, x i , x i+1 , x i+2 respectively represent the network parameters of the adjacent three time points before and after x, y i , y i+1 , y i+2 respectively represent the data values of the adjacent three time points before and after y.

[0048] Optionally, after obtaining the first historical network parameter, the method further comprises: respectively extracting features of the spatial information and the time information of the first historical network parameter to obtain spatial features and time features; weighting the spatial features and the time features by using an attention mechanism to obtain fusion features; inputting the fusion features into an initial model to train the initial model to obtain a first prediction model.

[0049] Specifically, the positional encoding method can be used to encode the time information to complete the feature extraction of the time information, and the multi head attention mechanism can be used to encode the spatial information.

[0050] In some embodiments of the present application, before the central server controls a plurality of remote servers to train the received first prediction model by using the second historical network parameters stored in the remote servers to obtain a plurality of second prediction models, the method further comprises: the central server sends a confirmation instruction to the plurality of remote servers to determine whether the first prediction model exists in the plurality of remote servers; in the case that the first prediction model is not received in the remote server, the central server sends the first prediction model to the remote server again which does not receive the first prediction model.

[0051] Specifically, the central server sends instructions to the remote servers corresponding to the distributed interaction machines. After receiving the instructions, the remote servers access the identification field in the historical database in the server. If the identification field is not detected, the remote servers send a request instruction through a program to request the central server to issue a model. After receiving the request instruction, the central server sends the model to the remote servers.

[0052] After the second prediction model is trained, the multiple remote servers can transmit data to the outside through an open platform. During the data transmission process, the central server controls the multiple remote servers to obtain network parameters in the data transmission process of the multiple remote servers. The central server controls the remote servers to transmit the network parameters to the remote prediction model. When the remote prediction model detects that there is an unverified parameter in the network parameters, a warning information is sent.

[0053] It should be noted that the unverified parameter includes a warning network parameter determined by the second prediction model as a parameter that will cause a failure in the data transmission process.

[0054] The embodiment of the application also provides another data transmission warning method, as shown in Figure 3 The method comprises the following steps:

[0055] In step S302, the remote server receives the first prediction model sent by the central server.

[0056] In step S304, the remote server trains the received first prediction model by using the second historical network parameter stored in the remote server to obtain a second prediction model.

[0057] In step S306, the remote server sends the trained second prediction model to the central server. The second prediction model is used to monitor the network parameters in the data transmission process of the remote server and to send a warning when the monitoring result meets a preset condition.

[0058] The data transmission warning method provided by the embodiment of the application is also applied to a data transmission warning device provided by the embodiment of the application, as shown in Figure 4As shown, comprising: a first training module 40, for training the initial model to obtain a first prediction model by using the first historical network parameters pre-stored in the central server; a sending module 42, for sending the first prediction model to a plurality of remote servers; a second training module 44, for controlling the plurality of remote servers to train the received first prediction model by using the second historical network parameters stored in the remote servers to obtain a plurality of second prediction models; a warning module 46, for receiving the plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are used to monitor the network parameters in the data transmission process of the plurality of remote servers, and in the case that the monitoring result meets the preset condition, a warning is given.

[0059] The first training module 40 comprises a training submodule, which is used to obtain the first historical network parameters, wherein the first historical network parameters at least include the resource occupancy rate at each sampling time in the first preset period and the network delay at each sampling time in the first preset period; and the network parameters at a plurality of times in the second preset period before the time of data transmission anomaly in the first preset period are taken as labels, and the network parameters at each sampling time in the first preset period are taken as training data set to train the first prediction model.

[0060] The training submodule comprises a data unit and an extraction unit, the data unit is used to obtain the first historical network parameters from the database in the central server; in the case that one or more time data is missing in the first historical network parameters, the missing data value is determined based on the data of the adjacent multiple times before and after the missing data by using the second difference method.

[0061] The extraction unit is used to respectively extract the spatial information and the time information of the first historical network parameters to obtain the spatial features and the time features; the spatial features and the time features are weighted by using the attention mechanism to obtain the fusion features; and the fusion features are input into the initial model to train the initial model to obtain the first prediction model.

[0062] The sending module 42 comprises an instruction submodule, which is used to send a confirmation instruction from the central server to the plurality of remote servers to determine whether the first prediction model exists in the plurality of remote servers; and in the case that the first prediction model is not received in the remote server, the central server sends the first prediction model to the remote server again which has not received the first prediction model.

[0063] The warning module 46 comprises a warning submodule, which is used to control the plurality of remote servers to obtain the network parameters in the data transmission process of the plurality of remote servers; control the remote servers to transmit the network parameters to the second prediction model, and when the second prediction model detects that there is an unverified parameter in the network parameters, a warning information is sent.

[0064] According to a further aspect of the embodiments of the present application, a non-transitory storage medium is also provided, including a stored program, wherein the program, when executed, controls a device in which the non-transitory storage medium is located to perform the data transmission early warning method.

[0065] According to a further aspect of the embodiments of the present application, an electronic device is also provided, including a memory and a processor, the processor being configured to execute a program, wherein the program, when executed, performs the data transmission early warning method.

[0066] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0067] In the above-mentioned embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0068] In the several embodiments of the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the unit embodiment described above is only schematic. For example, the division of the units can be a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, unit or module, and can be electrical or other forms.

[0069] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e. they can be located in one place, or distributed on multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0070] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0071] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0072] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be considered as the protection scope of the present application.

Claims

1. A data transmission early warning method, characterized in that, The method comprises the following steps: The central server trains an initial model by using first historical network parameters pre-stored in the central server to obtain a first prediction model. Specifically, the first historical network parameters are obtained, wherein the first historical network parameters at least include resource occupancy rate at each sampling time in a first preset time period and network delay at each sampling time in the first preset time period; network parameters at multiple time points in a second preset time period before a time point of data transmission anomaly in the first preset time period are taken as labels, and network parameters at each sampling time in the first preset time period are taken as training data set to train the first prediction model. After the first historical network parameters are obtained, spatial information and time information of the first historical network parameters are respectively extracted to obtain spatial features and time features; the spatial features and the time features are weighted by using an attention mechanism to obtain fusion features; and the fusion features are input into the initial model to train the initial model to obtain the first prediction model; The central server sends the first prediction model to multiple remote servers; The central server controls the multiple remote servers to train the received first prediction model by using second historical network parameters stored in the remote servers to obtain multiple second prediction models; The central server receives the multiple second prediction models trained by the multiple remote servers, wherein the multiple second prediction models are used to monitor network parameters in a data transmission process of the multiple remote servers, and perform early warning in a case where a monitoring result meets a preset condition.

2. The method of claim 1, wherein, The first historical network parameters are obtained by: Obtaining the first historical network parameters from a database in the central server; In a case where one or more time point data in the first historical network parameters are missing, the missing data values are determined based on data of adjacent multiple time points before and after the missing data by using a second difference method.

3. The method of claim 1, wherein, Before the central server controls the multiple remote servers to train the received first prediction model by using second historical network parameters stored in the remote servers to obtain multiple second prediction models, the method further comprises: The central server sends a confirmation instruction to the multiple remote servers to determine whether the first prediction model exists in the multiple remote servers; In a case where the first prediction model is not received in the remote servers, the central server sends the first prediction model to the remote servers again which do not receive the first prediction model.

4. The method of claim 1, wherein, The multiple second prediction models are used to monitor network parameters in a data transmission process of the multiple remote servers, and perform early warning in a case where a monitoring result meets a preset condition, which comprises: The central server controls the multiple remote servers to obtain network parameters in a data transmission process of the multiple remote servers; The central server controls the remote server to transmit the network parameters into the second prediction model, and sends a warning message when the second prediction model detects that there is an unverified parameter in the network parameters.

5. A data transmission early warning method, characterized in that, Comprise: The remote server receives a first prediction model sent by the central server, wherein the central server obtains first historical network parameters, wherein the first historical network parameters at least include resource occupancy rate at each sampling time in a first preset period and network delay at each sampling time in the first preset period; network parameters at multiple time points in a second preset period before the time point of data transmission anomaly in the first preset period are taken as labels, and network parameters at each sampling time in the first preset period are taken as training data set to train the first prediction model; after obtaining the first historical network parameters, spatial information and time information of the first historical network parameters are respectively extracted to obtain spatial features and time features; the spatial features and the time features are weighted by using an attention mechanism to obtain fusion features; the fusion features are input into an initial model to train the initial model to obtain the first prediction model; The remote server trains the received first prediction model by using second historical network parameters stored in the remote server to obtain a second prediction model; The remote server sends the trained second prediction model to the central server, wherein the second prediction model is used to monitor network parameters in the data transmission process of the remote server, and sends a warning message when the monitoring result meets a preset condition.

6. A data transmission early warning device, characterized in that Comprise: The first training module is used for training an initial model by using first historical network parameters pre-stored in the central server to obtain a first prediction model, specifically, the first training module obtains the first historical network parameters, wherein the first historical network parameters at least include resource occupancy rate at each sampling time in a first preset period and network delay at each sampling time in the first preset period; network parameters at multiple time points in a second preset period before the time point of data transmission anomaly in the first preset period are taken as labels, and network parameters at each sampling time in the first preset period are taken as training data set to train the first prediction model, wherein after obtaining the first historical network parameters, spatial information and time information of the first historical network parameters are respectively extracted to obtain spatial features and time features; the spatial features and the time features are weighted by using an attention mechanism to obtain fusion features; the fusion features are input into the initial model to train the initial model to obtain the first prediction model; The sending module is used for sending the first prediction model to multiple remote servers; The second training module is used for controlling the multiple remote servers to train the received first prediction model by using second historical network parameters stored in the remote server to obtain multiple second prediction models; An early warning module is configured to receive a plurality of second prediction models trained by the plurality of remote servers, wherein the plurality of second prediction models are configured to monitor network parameters in a data transmission process of the plurality of remote servers, and perform early warning when a monitoring result meets a preset condition.

7. A non-volatile storage medium, characterized by The non-volatile storage medium comprises a stored program, wherein the non-volatile storage medium controls a device in which the non-volatile storage medium is located to perform the data transmission early warning method of any one of claims 1 to 5 when the program is running.

8. An electronic device, comprising: A device comprises a memory and a processor, wherein the processor is configured to run a program, and the device performs the data transmission early warning method of any one of claims 1 to 5 when the program is running.

Images