Operation and maintenance log storage method and device

Abstract

The application provides an operation and maintenance log storage method and device, which can be used in the financial field or other fields. The method comprises the following steps: acquiring real-time operation and maintenance data collected by an operation and maintenance system; analyzing the real-time operation and maintenance data by using a storage type decision model constructed by a neural network algorithm to obtain a storage type result; and storing and processing operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type result. The storage type decision model is obtained by training the neural network algorithm, so that the storage type of the operation and maintenance log is determined. The problem that a large amount of redundant information is generated in the system log recorded by the existing system under a stable operation state, a large amount of storage space is wasted, and the search performance of the log content is reduced is solved. More operation and maintenance index information can be stored by using relatively less storage space, a large amount of redundant data is avoided, and the storage space is saved.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance log technology, and can be applied to the financial field and other fields, particularly to an operation and maintenance log storage method and device. Background Technology

[0002] Currently, IT operations and maintenance management is one of the hottest topics in the information technology field. With the continuous deepening and improvement of IT infrastructure, the operation and maintenance of computer hardware and software systems has become a widespread and overwhelming problem for information service departments across various industries. As data volume rapidly increases, the number of devices continues to rise, and hardware and software systems become increasingly complex, the indicators that various operations and maintenance monitoring systems need to monitor are also increasing. Consequently, the content of logs recording relevant information and the storage space they occupy are both continuously and rapidly increasing.

[0003] It can be seen that as the application scenarios deepen and the operation and maintenance system is continuously built, some bottlenecks will be encountered. For example, when some systems are relatively stable, some system indicators are also relatively stable, and even remain unchanged at a certain value for a long period of time, or the indicator data undergoes periodic changes; however, this will result in a large amount of duplicate information in the recorded logs, generating a large amount of data redundancy, wasting a lot of storage space, and also causing performance and time losses when searching logs. Summary of the Invention

[0004] In view of the problems existing in the prior art, the main purpose of the embodiments of this application is to provide an intelligent operation and maintenance method and device to avoid generating a large amount of redundant data and save storage space.

[0005] To achieve the above objectives, this application provides a method for storing operation and maintenance logs. The method includes: acquiring real-time operation and maintenance data collected by the operation and maintenance system; analyzing the real-time operation and maintenance data using a storage type decision model constructed through a neural network algorithm to obtain a storage type result; and storing the operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type result.

[0006] In the above-mentioned operation and maintenance log storage method, optionally, the storage type result obtained by analyzing the real-time operation and maintenance data through the storage type decision model constructed by the neural network algorithm includes: inputting the real-time operation and maintenance data into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data; and generating the storage type result corresponding to the real-time operation and maintenance data according to the preset storage rules and the time interval information.

[0007] In the above-mentioned operation and maintenance log storage method, optionally, the storage processing of the operation and maintenance log corresponding to the real-time operation and maintenance data according to the storage type result includes: obtaining the data change status according to the time interval information in the storage type result; storing the real-time operation and maintenance data completely or storing a segment according to the data change status, and recording the corresponding start and end time points; wherein, the time interval information includes the time interval type and the start and end time points.

[0008] In the above-mentioned operation and maintenance log storage method, optionally, storing the real-time operation and maintenance data completely or storing fragments according to the data change status, and recording the corresponding start and end time points includes: when the change frequency of the data change status is lower than a preset threshold, performing a discrete Fourier transform on the real-time operation and maintenance data to obtain frequency information, extracting target data in the frequency information that is lower than a preset frequency band threshold for storage, and recording the start and end time points of the corresponding time interval; when the change frequency of the data change status is equal to or higher than the preset threshold, storing the real-time operation and maintenance data completely and recording the start and end time points of the fault occurrence process.

[0009] Optionally, in the above-mentioned operation and maintenance log storage method, the method further includes: acquiring historical system operation and maintenance data, training an initial fault model using a neural network algorithm based on the historical system operation and maintenance data, and using a Transformer algorithm to evolve the parameters of the initial storage type model to obtain the storage type decision model.

[0010] In the above-mentioned operation and maintenance log storage method, optionally, the step of training the initial storage type model using a neural network algorithm based on the historical system operation and maintenance data includes: using a graph convolutional neural network algorithm to extract features from the historical system operation and maintenance data to obtain a historical data feature matrix; and using the historical data feature matrix to train the initial storage type model to obtain a trained initial storage type model.

[0011] In the above-mentioned operation and maintenance log storage method, optionally, the storage type decision model is obtained by using the Transformer algorithm to evolve the parameters of the initial storage type model, which includes: using the encoder in the Transformer algorithm to perform normalization and nonlinear transformation processing on the trained initial storage type model; and using the decoder in the Transformer algorithm to decode the model processed by the encoder to obtain the storage type decision model.

[0012] This application also provides an operation and maintenance log storage device, the device comprising: a collection module for acquiring real-time operation and maintenance data collected by the operation and maintenance system; an analysis module for analyzing the real-time operation and maintenance data to obtain storage type results through a storage type decision model constructed by a neural network algorithm; and a processing module for storing and processing the operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type results.

[0013] Optionally, in the above-mentioned operation and maintenance log storage device, the device may further include a model training module. The model training module is used to acquire historical system operation and maintenance data, train an initial fault model using a neural network algorithm based on the historical system operation and maintenance data, and evolve the parameters of the initial storage type model using the Transformer algorithm to obtain the storage type decision model.

[0014] Optionally, the above-mentioned operation and maintenance log storage device may further include a preprocessing module, which is used to preprocess historical system operation and maintenance data; wherein, the preprocessing includes noise filtering, data cleaning and data analysis.

[0015] In the aforementioned operation and maintenance log storage device, optional historical system operation and maintenance data may include: system success rate, CPU utilization, memory utilization, service response time, service consumption time, network speed, and monitoring messages.

[0016] In the aforementioned operation and maintenance log storage device, the optional analysis module includes:

[0017] The time interval unit is used to input real-time operation and maintenance data into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data;

[0018] The storage type unit is used to generate the storage type result corresponding to the real-time operation and maintenance data based on the preset storage rules and the time interval information corresponding to the real-time operation and maintenance data.

[0019] Optionally, in one embodiment of this application, the time interval information includes: time interval type and start and end times.

[0020] Optionally, in one embodiment of this application, the device further includes: a feedback information module, used to acquire and store log storage feedback information, and to update the storage type decision model using the log storage feedback information.

[0021] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method.

[0022] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0023] This application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the above-described method.

[0024] This application trains a storage type decision model using a neural network algorithm to determine the storage type of operation and maintenance logs. This solves the problem that existing systems generate a large amount of redundant information in their logs under stable operating conditions, resulting in a significant waste of storage space and performance degradation in log content search. It allows for storing more operation and maintenance indicator information with relatively less storage space, avoiding the generation of a large amount of redundant data and saving storage space. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart illustrating an operation and maintenance log storage method according to an embodiment of this application;

[0027] Figure 2 This is a flowchart illustrating the generation of storage type results in the embodiments of this application;

[0028] Figure 3 This is a schematic diagram of the system architecture for applying the operation and maintenance log storage method in the embodiments of this application;

[0029] Figure 4 This is a system workflow diagram in the embodiments of this application;

[0030] Figure 5 This is a schematic diagram of the structure of an operation and maintenance log storage device according to an embodiment of this application;

[0031] Figure 6 This is a schematic diagram of the structure of the operation and maintenance log storage device in another embodiment of this application;

[0032] Figure 7 This is a schematic diagram of the storage type module in an embodiment of this application;

[0033] Figure 8 This is a schematic diagram of the structure of the operation and maintenance log storage device in another embodiment of this application;

[0034] Figure 9This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0035] This application provides an intelligent operation and maintenance method and apparatus, which can be used in the financial field and other fields. It should be noted that the intelligent operation and maintenance method and apparatus of this application can be used in the financial field, or in any field other than the financial field. The application field of the intelligent operation and maintenance method and apparatus of this application is not limited.

[0036] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0037] like Figure 1 The diagram shows a flowchart of an operation and maintenance log storage method according to an embodiment of this application. The execution subject of the operation and maintenance log storage method provided in this embodiment includes, but is not limited to, a computer. This application obtains a storage type decision model through neural network algorithm training, thereby determining the storage type of operation and maintenance logs. This solves the problem that existing systems generate a large amount of redundant information in the system logs under stable operating conditions, resulting in a large waste of storage space and performance loss in log content search. More operation and maintenance indicator information can be stored in relatively less storage space, avoiding the generation of a large amount of redundant data and saving storage space. Figure 1 The methods shown include:

[0038] Step S1: Obtain real-time operation and maintenance data collected by the operation and maintenance system;

[0039] Step S2: Analyze the real-time operation and maintenance data using a storage type decision model constructed through a neural network algorithm to obtain storage type results;

[0040] Step S3: Store the operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type result.

[0041] In the above embodiments, various information system operation index data are acquired. This type of index data is updated over time and recorded in the log, referred to as "time-series operation and maintenance data". This application uses a model trained by a neural network to intelligently segment the "time-series operation and maintenance data" within a set time period, dividing the time period into stable or even periodic time intervals, or unstable and irregular time intervals when the system fails. Different storage methods are used for the "time-series operation and maintenance data" in different types of time intervals.

[0042] As one embodiment of this application, such as Figure 2 As shown, the storage type decision model constructed using a neural network algorithm analyzes the real-time operation and maintenance data to obtain storage type results, including:

[0043] Step S21: Input the real-time operation and maintenance data into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data;

[0044] Step S22: Generate the storage type result corresponding to the real-time operation and maintenance data according to the preset storage rules and the time interval information.

[0045] In this embodiment, the time interval information includes: time interval type and start and end times.

[0046] Specifically, the data change status is obtained based on the time interval information in the storage type result; the real-time operation and maintenance data is then stored completely or in fragments based on the data change status, and the corresponding start and end times are recorded. In this embodiment, the real-time operation and maintenance data is mainly determined based on its change status to determine whether it has reference value. If there is no fault, the real-time operation and maintenance data at this stage is considered routine operation data and does not need to be saved in its entirety. Therefore, saving a portion of the data can reduce the waste of storage space. Conversely, if a fault exists, the entire data must be saved.

[0047] Furthermore, based on the data change state, the real-time operation and maintenance data is fully stored or fragmented for storage, and the corresponding start and end time points are recorded, including: when the change frequency of the data change state is lower than a preset threshold, the real-time operation and maintenance data is subjected to discrete Fourier transform to obtain frequency information, and target data below the preset frequency band threshold in the frequency information is extracted, stored, and the start and end time points of the corresponding time interval are recorded; when the change frequency of the data change state is equal to or higher than the preset threshold, the real-time operation and maintenance data is fully stored and the start and end time points of the fault occurrence process are recorded.

[0048] Specifically, the time interval information corresponding to real-time operation and maintenance data is defined as a time interval that is stable or even periodic within a certain set length, or unstable and irregular when the system fails. The stability can be judged by comparing the change frequency with the preset threshold. In practice, those skilled in the art can also use other methods to distinguish the changes, such as the slope and variance after fitting. This application does not make any further limitations here.

[0049] Furthermore, the preset storage rules include the storage types corresponding to operation and maintenance logs in different time intervals. For example, the storage rule for unstable operation and maintenance logs when the system fails can be to not store them.

[0050] Therefore, based on the preset storage rules, the storage type of the real-time operation and maintenance data is determined by utilizing the time interval information corresponding to the real-time operation and maintenance data.

[0051] The storage type result indicates the storage method corresponding to different real-time operation and maintenance data. Operation and maintenance logs corresponding to the real-time operation and maintenance data are stored according to the storage type result. This allows for different processing and storage methods to be used for operation and maintenance data of different storage types, effectively avoiding wasted storage space and improving the efficiency of operation and maintenance log storage.

[0052] In one embodiment of this application, the method further includes acquiring historical system operation and maintenance data, training an initial fault model using a neural network algorithm based on the historical system operation and maintenance data, and evolving the parameters of the initial storage type model using a Transformer algorithm to obtain the storage type decision model.

[0053] Specifically, historical system operation and maintenance data may include: system success rate, CPU utilization, memory utilization, service response time, service latency, network speed, and monitoring messages. In practice, historical system operation and maintenance data of IT systems are collected, including business success rate, system success rate, CPU utilization, memory utilization, service response time, service latency, network speed, monitoring messages, various system-level and application-level logs generated during system operation, as well as accumulated raw alarm information and corresponding manually annotated alarm information, etc.

[0054] As an embodiment of this application, the method further includes: preprocessing historical system operation and maintenance data; wherein, the preprocessing includes noise filtering, data cleaning, and data analysis; wherein, noise filtering, data cleaning, data correlation analysis, and principal component analysis are performed on the historical system operation and maintenance data to ensure the accuracy of the data.

[0055] Furthermore, the preprocessed historical system operation and maintenance data is used as training sample data for the model to train a log storage type decision model based on a neural network. Using neural network technology, the initial storage type model is trained based on the historical system operation and maintenance data to obtain a log storage type decision model based on a neural network. This storage type decision model can be used to divide a set period of time into stable, even periodic, or unstable and irregular time intervals when the system experiences failures. Ultimately, the type of the divided time interval, its start and end times, and the complete operation and maintenance data within that time interval are obtained.

[0056] In one embodiment of this application, the method of training an initial storage type model using a neural network algorithm based on historical system operation and maintenance data to obtain a storage type decision model includes: using a graph convolutional neural network algorithm to extract features from the historical system operation and maintenance data to obtain a historical data feature matrix; and using the historical data feature matrix to train the initial storage type model to obtain a trained initial storage type model.

[0057] The storage type decision model for intelligent time interval partitioning is implemented using Transformer. The Transformer algorithm consists of an encoder and a decoder.

[0058] Furthermore, the encoder consists of N = 6 identical layers, with multiple layers represented as "Nx", where x6 is the number. Each layer consists of two sub-layers: a multi-head self-attention mechanism and a fully connected feed-forward network.

[0059] Each intermediate layer has a residual connection and normalization, so the output of the intermediate layer can be represented as shown in Equation (1).

[0060] sub_layer_output=LayerNorm(x+(SubLayer(x))) (1)

[0061] Among them, there are two intermediate layers:

[0062] 1) Bullish Self-Attention

[0063] The attention mechanism can be represented by formula (2).

[0064] attemtopm_output=Attention(Q,k,v) (2)

[0065] Among them, multi-head self-attention projects Q, K, and V through h different linear transformations, and finally splices the different attention results together, as shown in formula (3) and formula (4).

[0066] MultiHead(Q,K,V)=Concat(head1,...,head h W O(3)

[0067]

[0068] In this case, self-attention takes Q, K, and V as the same.

[0069] In addition, the attention calculation uses a scaled dot-product, as shown in Equation (5).

[0070]

[0071] 2) Position-wise feed-forward networks

[0072] This layer primarily provides non-linear transformations. The dimension of the Attention output is [bsz*seq_len, num_heads*head_size]. The second intermediate layer is a fully connected layer. It is position-wise because the transformation parameters at each position i are the same when passing through the linear layer.

[0073] The decoder and encoder have similar structures, but the decoder has an additional attention layer. Let's first clarify the input, output, and decoding process of the decoder:

[0074] (1) Output: The probability distribution of the output word at position i;

[0075] (2) Input: The output of the encoder & the output of the decoder at position i-1. Therefore, the attention in the middle is not self-attention; its K and V come from the encoder, and its Q comes from the output of the decoder at the previous position.

[0076] (3) Decoding: Encoding can be computed in parallel and all sequences can be encoded at once, but decoding is not done by solving all sequences at once, but by solving them one by one like an RNN, because the input of the previous position is used as the query for attention.

[0077] Furthermore, for stable or even periodic maintenance data within a defined time interval, the system uses Discrete Fourier Transform and a low-pass filter to extract the low-frequency portion of the frequency domain information. Finally, it stores the time interval type, start and end times, and the extracted frequency domain information. For maintenance data that is unstable or irregularly changing during system failures, the system stores the time interval type, start and end times, and complete maintenance data information within the interval.

[0078] The methods for using Discrete Fourier Transform and low-pass filtering are described below:

[0079] The formula for the Discrete Fourier Transform is shown in formula (6).

[0080]

[0081] Where 0≤k≤N-1. The formula for the low-pass filter is shown in formula (7).

[0082]

[0083] As an embodiment of this application, the method further includes: acquiring and storing log storage feedback information, and using the log storage feedback information to update the storage type decision model.

[0084] This involves acquiring and storing log storage feedback information input by operations and maintenance personnel. This log storage feedback information is then used to optimize and update the storage type decision model, thereby continuously improving its accuracy.

[0085] Specifically, for example, the system uses Discrete Fourier Transform to obtain frequency domain information for operational data within stable or even periodic intervals, extracting and storing only low-frequency information, while simultaneously recording the start and end times of the corresponding time interval. This significantly reduces the amount of redundant information stored, and such data can be converted from the frequency domain to the time domain using Inverse Discrete Fourier Transform for normal information retrieval. The system completely preserves operational data within time intervals when the data is unstable or even experiencing failures. Therefore, this application avoids storing a large amount of redundant data generated during stable system operation and completes the storage of detailed operational indicator data when the system is unstable or experiencing failures.

[0086] In this embodiment, as Figure 3 The diagram shown is a schematic representation of the system architecture for applying the operation and maintenance log storage method in this embodiment of the application, specifically including:

[0087] Operations and Maintenance Information Acquisition and Processing Unit: This unit collects various information data during the operation of the IT system and preprocesses it before passing it to the AIOPS model unit. This unit is responsible for collecting data such as business success rate, system success rate, CPU utilization, memory utilization, service response time, service latency, network speed, monitoring messages, various system-level and application-level logs generated during system operation, as well as accumulated raw alarm information and corresponding manually labeled alarm information. Furthermore, this module performs data preprocessing such as noise filtering, data cleaning, data correlation analysis, principal component analysis, and PCA dimensionality reduction on this information.

[0088] AIOPS Model Unit: After receiving the data from the Operation and Maintenance Information Acquisition and Processing Unit, the AIOPS Model Unit uses a neural network-based log storage type decision model to intelligently divide the time storage interval of "time-series operation and maintenance data", and then transmits the start and end times of the final divided time interval and the complete operation and maintenance data within that time period to the Log Intelligent Storage Unit.

[0089] Log Intelligent Storage Unit: After receiving data from the AIOPS model unit, the log intelligent storage unit processes and stores different types of data using different methods. Log information requiring complete preservation is stored in its entirety, while log information not requiring complete preservation is processed using Discrete Fourier Transform and low-pass filtering before storage.

[0090] Raw log reading unit: The raw log reading unit reads log information with complete information directly, and reads log information with incomplete information after processing with discrete Fourier inverse transform, depending on the different storage methods of the logs in the intelligent log storage unit.

[0091] AIOPS, or Artificial Intelligence for IT Operations, applies artificial intelligence to the field of operations and maintenance. Based on existing operational data (logs, monitoring information, application information, etc.), it uses machine learning, deep learning, and other methods to further solve problems that traditional operations and maintenance methods cannot address. AIOPS does not rely on manually specified rules; it advocates that artificial intelligence algorithms automatically learn from massive amounts of operational data (including the events themselves, various types of information data, and logs manually processed by operations personnel), continuously refining and summarizing rules.

[0092] Furthermore, the Fourier Transform represents the ability to express a function that satisfies certain conditions as a linear combination of trigonometric functions (sine or cosine functions) or their integrals. In different research fields, the Fourier Transform has various variants, such as the continuous Fourier Transform and the discrete Fourier Transform.

[0093] The Discrete Fourier Transform (DFT) is designed for use in scientific computing and digital signal processing to perform Fourier transforms on computers, making the Fourier transform discrete in both the time and frequency domains. It transforms the sampling of time-domain signals into sampling in the frequency domain of the Discrete-Time Fourier Transform (DTFT).

[0094] Specifically, in form, the sequences at both ends of the Discrete Fourier Transform (in the time and frequency domains) are of finite length, but in reality, both sets of sequences should be considered as principal value sequences of discrete periodic signals. Even when performing a DFT on a finite-length discrete signal, it should be regarded as a periodically extended signal before the transformation.

[0095] In this embodiment, the system of this application uses a neural network to intelligently divide the "time-series operation and maintenance data" into time intervals, and performs subsequent data processing and storage as follows: Figure 4 As shown. Specifically:

[0096] Step 1: Acquisition of Operational and Maintenance Information. Acquire various operational and maintenance data accumulated during the historical operation of the IT system. This data includes: business success rate, system success rate, CPU utilization, memory utilization, service response time, service latency, network speed, monitoring messages, original alarm information, operations and feedback information from operations and maintenance personnel, and various logs generated by the system, along with corresponding annotations indicating whether the system is operating normally. Perform noise filtering, data cleaning, data correlation analysis, principal component analysis, and PCA dimensionality reduction on this data for preprocessing. Then, use this raw data and the preprocessed, annotated data to construct an IT system operational and maintenance information database.

[0097] Step 2: Train a neural network-based log storage type decision model. Using neural network technology, train the model based on the IT system operation and maintenance information database obtained in Step 1 to obtain a neural network-based log storage type decision model.

[0098] Among them, the log storage type decision model can divide a time period into stable, even periodic, or unstable and irregular time intervals when the system fails, within a certain set length of time; and finally obtain the final divided time interval type, start and end time points, and complete operation and maintenance data within that time period.

[0099] Step 3: Real-time / Batch Decision Making. Based on the model in Step 2, make real-time or batch decisions on real-time or historical operation and maintenance data. Finally, divide the real-time / historical operation and maintenance data into start and end time intervals and provide complete operation and maintenance data for the corresponding time periods.

[0100] Step Four: Processing and Storage of Operation and Maintenance Data Logs. Based on the start and end times of the time intervals defined in Step Three and the complete operation and maintenance data within those time periods, different processing and storage methods are used for different types of data. Furthermore, different reading methods are employed for logs stored in different ways when reading raw log information.

[0101] This application trains a storage type decision model using a neural network algorithm to determine the storage type of operation and maintenance logs. This solves the problem that existing systems generate a large amount of redundant information in their logs under stable operating conditions, resulting in a significant waste of storage space and performance degradation in log content search. It allows for storing more operation and maintenance indicator information with relatively less storage space, avoiding the generation of a large amount of redundant data and saving storage space.

[0102] like Figure 5 The figure shows a schematic diagram of an operation and maintenance log storage device according to an embodiment of this application. The device shown in the figure includes:

[0103] The acquisition module 10 is used to acquire real-time operation and maintenance data collected by the operation and maintenance system;

[0104] Analysis module 20 is used to analyze the real-time operation and maintenance data to obtain storage type results through a storage type decision model constructed by a neural network algorithm;

[0105] The processing module 30 is used to store the operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type result.

[0106] In the above embodiments, the device may further include a model training module, which is used to acquire historical system operation and maintenance data, train an initial fault model using a neural network algorithm based on the historical system operation and maintenance data, and evolve the parameters of the initial storage type model using a Transformer algorithm to obtain the storage type decision model. Further, in one embodiment of this application, as... Figure 6 As shown, the device may further include a preprocessing module 40 for preprocessing historical system operation and maintenance data; wherein, the preprocessing includes noise filtering, data cleaning, and data analysis. The historical system operation and maintenance data includes: system success rate, CPU utilization, memory utilization, service response time, service latency, network speed, and monitoring messages.

[0107] Optionally, in one embodiment of this application, such as Figure 7 As shown, the storage type module 20 includes:

[0108] Time interval unit 21 is used to input real-time operation and maintenance data into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data;

[0109] Storage type unit 22 is used to generate storage type results corresponding to real-time operation and maintenance data based on preset storage rules and time interval information corresponding to real-time operation and maintenance data.

[0110] Optionally, in one embodiment of this application, the time interval information includes: time interval type and start and end times.

[0111] Optionally, in one embodiment of this application, such as Figure 8 As shown, the device also includes a feedback information module 50, used to acquire and store log storage feedback information, and to update the storage type decision model using the log storage feedback information.

[0112] Based on the same concept as the aforementioned operation and maintenance log storage method, this application also provides an operation and maintenance log storage device. Since the principle by which this operation and maintenance log storage device solves the problem is similar to that of the operation and maintenance log storage method, the implementation of this operation and maintenance log storage device can refer to the implementation of the operation and maintenance log storage method; details that are repeated will not be repeated.

[0113] This application trains a storage type decision model using a neural network algorithm to determine the storage type of operation and maintenance logs. This solves the problem that existing systems generate a large amount of redundant information in their logs under stable operating conditions, resulting in a significant waste of storage space and performance degradation in log content search. It allows for storing more operation and maintenance indicator information with relatively less storage space, avoiding the generation of a large amount of redundant data and saving storage space.

[0114] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method.

[0115] This application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the above-described method.

[0116] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0117] like Figure 9 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 9 All components shown; in addition, the electronic device 600 may also include Figure 9 For components not shown, please refer to existing technologies.

[0118] like Figure 9 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0119] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.

[0120] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0121] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.

[0122] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0123] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0124] Based on different communication technologies, multiple communication modules 110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling typical telecommunications functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 130 is coupled to a central processing unit 100, enabling on-device recording via the microphone 132 and on-device playback of stored audio via the speaker 131.

[0125] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0127] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0129] This application uses specific embodiments to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for storing operation and maintenance logs, characterized in that, The method includes: Obtain real-time operation and maintenance data collected by the operation and maintenance system; The storage type decision model, constructed using a neural network algorithm, analyzes the real-time operation and maintenance data to obtain storage type results. The operation and maintenance logs corresponding to the real-time operation and maintenance data are stored and processed according to the storage type result. Acquire and store log storage feedback information, and use the log storage feedback information to update the storage type decision model; The storage type decision model constructed using a neural network algorithm analyzes the real-time operation and maintenance data to obtain storage type results, including: The real-time operation and maintenance data is input into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data; wherein, the time interval information includes the time interval type and start and end time points; The storage type result corresponding to the real-time operation and maintenance data is generated according to the preset storage rules and the time interval information; The storage processing of the operation and maintenance logs corresponding to the real-time operation and maintenance data based on the storage type result includes: The data change status is obtained based on the time interval information in the storage type result; When the frequency of the data change state is lower than a preset threshold, the real-time operation and maintenance data is subjected to discrete Fourier transform to obtain frequency information, and the target data in the frequency information that is lower than the preset frequency band threshold is extracted, stored, and the start and end time points of the corresponding time interval are recorded. When the frequency of the data change state is equal to or higher than a preset threshold, the real-time operation and maintenance data will be completely stored and the start and end times of the fault occurrence process will be recorded.

2. The operation and maintenance log storage method according to claim 1, characterized in that, The method further includes: Historical system operation and maintenance data is acquired. Based on the historical system operation and maintenance data, a neural network algorithm is used to train an initial fault model. The parameters of the initial storage type model are then evolved using the Transformer algorithm to obtain the storage type decision model.

3. The operation and maintenance log storage method according to claim 2, characterized in that, The step of training the initial storage type model using a neural network algorithm based on the historical system operation and maintenance data includes: The historical system operation and maintenance data are used to extract features from the historical data using a graph convolutional neural network algorithm to obtain a historical data feature matrix. The initial storage type model is trained using the historical data feature matrix to obtain the trained initial storage type model.

4. The operation and maintenance log storage method according to claim 3, characterized in that, The parameters of the initial storage type model are evolved using the Transformer algorithm to obtain the storage type decision model, which includes: The encoder in the Transformer algorithm is used to perform normalization and nonlinear transformation on the trained initial storage type model. The storage type decision model is obtained by decoding the model processed by the encoder using the decoder in the Transformer algorithm.

5. A maintenance log storage device, characterized in that, The device includes: The data acquisition module is used to acquire real-time operation and maintenance data collected by the operation and maintenance system. The analysis module is used to analyze the real-time operation and maintenance data to obtain storage type results through a storage type decision model constructed using a neural network algorithm; The processing module is used to store and process the operation and maintenance logs corresponding to the real-time operation and maintenance data according to the storage type result; The feedback information module is used to acquire and store log storage feedback information, and to update the storage type decision model using the log storage feedback information. The storage type module includes: A time interval unit is used to input the real-time operation and maintenance data into the storage type decision model to obtain the time interval information corresponding to the real-time operation and maintenance data; wherein, the time interval information includes the time interval type and start and end time points; The storage type unit is used to generate the storage type result corresponding to the real-time operation and maintenance data according to the preset storage rules and the time interval information. The processing module is specifically used for: The data change status is obtained based on the time interval information in the storage type result; When the frequency of the data change state is lower than a preset threshold, the real-time operation and maintenance data is subjected to discrete Fourier transform to obtain frequency information, and the target data in the frequency information that is lower than the preset frequency band threshold is extracted, stored, and the start and end time points of the corresponding time interval are recorded. When the frequency of the data change state is equal to or higher than a preset threshold, the real-time operation and maintenance data will be completely stored and the start and end times of the fault occurrence process will be recorded.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that performs the method according to any one of claims 1 to 4.

Images