Offline transaction control method, state prediction model training method and device

Abstract

The present application provides an offline transaction control method, a state prediction model training method, and an apparatus, which are applied to the field of computer technology. The method comprises: if an offline transaction in an online running environment is in a running state, obtaining an online running log corresponding to the online running environment within a preset historical period before a current moment; extracting transaction state data and device resource data from the online log data, integrating the transaction state data and the device resource data to obtain a formatted running log; inputting the formatted running log into a pre-trained state prediction model to obtain a prediction result of the environmental state of the online running environment caused by the offline transaction; obtaining a state intervention strategy corresponding to the prediction result, and adjusting the running state of the offline transaction based on the state intervention strategy, thereby improving the environmental stability of the online running environment.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to an offline transaction control method, a state prediction model training method and device. Background Art

[0002] Traditional offline transaction management relies on static rules and manual adjustments. Static rules mean that established planning management guidelines lack flexibility and are difficult to update in real time based on changing circumstances. Manual adjustments, on the other hand, rely too heavily on manual labor, resulting in low efficiency and prone to human error. In batch planning scenarios, high machine loads can lead to slower processing speeds and increased risk of failure. Low interface success rates can easily lead to data transmission interruptions, impacting the smooth progress of the entire plan. Summary of the Invention

[0003] This specification provides an offline transaction control method, a state prediction model training method, and a device, aiming to optimize the control process of offline transactions and thereby enhance the user experience. The technical solution is as follows:

[0004] In a first aspect, an embodiment of this specification provides an offline transaction control method, the method comprising:

[0005] If the offline transaction in the online running environment is in the running state, then obtain the online running log corresponding to the online running environment within the preset historical length before the current moment;

[0006] Extracting transaction status data and device resource data from online log data, integrating the transaction status data and device resource data to obtain a formatted operation log;

[0007] Input the formatted operation log into the pre-trained state prediction model to obtain the prediction results of the environmental state of the online operation environment caused by the offline transaction;

[0008] Obtain the status intervention strategy corresponding to the prediction result, and adjust the running status of the offline transaction based on the status intervention strategy.

[0009] In a second aspect, the embodiments of this specification provide a state prediction model training method, including:

[0010] Obtain preset training samples and test samples. The training samples are the actual operating states corresponding to the online operating environment under different transaction status data and device resource data. The test samples are composed of multiple different transaction status data and device resource data.

[0011] Input the training samples into the state prediction model to obtain the predicted operating state, and call the mean square error loss function to calculate the state error between the predicted operating state and the actual operating state;

[0012] Call the back propagation algorithm to adjust the model parameters of the state prediction model based on the state error so that the mean square error loss function reaches a convergence state;

[0013] The state prediction model is tested based on the test sample to determine the training state of the state prediction model.

[0014] In a third aspect, an embodiment of this specification provides a device for controlling offline transactions, including:

[0015] Log acquisition unit: used for acquiring the online operation log corresponding to the online operation environment within a preset historical period before the current moment if the offline transaction in the online operation environment is in the running state;

[0016] Data processing unit: used to extract transaction status data and device resource data from online log data, integrate the transaction status data and device resource data, and obtain formatted operation logs;

[0017] Result prediction unit: used to input formatted operation logs into a pre-trained state prediction model to obtain the prediction results of the offline transaction on the environmental state of the online operation environment;

[0018] State adjustment unit: used to obtain the state intervention strategy corresponding to the prediction result and adjust the running state of the offline transaction based on the state intervention strategy.

[0019] In a fourth aspect, an embodiment of this specification provides a training device for a state prediction model, comprising:

[0020] A sample acquisition unit is used to acquire preset training samples and test samples. The training samples are the actual operating states corresponding to the online operating environment under different transaction status data and device resource data. The test samples are composed of multiple different transaction status data and device resource data.

[0021] A sample processing unit is used to input the training sample into the state prediction model to obtain the predicted operating state, and call the mean square error loss function to calculate the state error between the predicted operating state and the actual operating state;

[0022] A parameter adjustment unit, used to call the back propagation algorithm to adjust the model parameters of the state prediction model based on the state error so that the mean square error loss function reaches a convergence state;

[0023] The state determination unit is used to test the state prediction model based on the test sample to determine the training state of the state prediction model.

[0024] In a fifth aspect, an embodiment of this specification provides an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements any of the above methods when executed by the processor.

[0025] In a fourth aspect, an embodiment of this specification provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, any of the above methods is implemented.

[0026] In the above technical solution, by obtaining online operation logs within a preset historical period, it is possible to fully understand the operating status and historical trends of the online environment, providing a data basis for subsequent analysis. Secondly, by extracting transaction status data and device resource data from the logs and integrating them, a formatted operation log can be generated to ensure the structured and analyzable nature of the data. Next, the formatted log is input into a pre-trained state prediction model, which can accurately predict the impact of offline transactions on the state of the online environment, thereby identifying potential problems in advance. Finally, based on the prediction results, the corresponding state intervention strategy is obtained, which can dynamically adjust the operating status of offline transactions, avoid transaction conflicts, and ensure the stable operation of the online operating environment. This process not only realizes the intelligent management of offline transactions, but also significantly reduces the cost of manual intervention and improves the environmental stability of the online operating environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 This is a system architecture diagram of an offline transaction control method provided by an embodiment of this specification;

[0029] Figure 2 This is a flowchart of an offline transaction control method provided by an embodiment of this specification;

[0030] Figure 3 This is a flowchart of an offline transaction control method provided by an embodiment of this specification;

[0031] Figure 4 This is a flowchart of an offline transaction control method provided by an embodiment of this specification;

[0032] Figure 5 This is a flow chart of a state prediction model training method provided in an embodiment of this specification;

[0033] Figure 6 This is a flow chart of a state prediction model training method provided in an embodiment of this specification;

[0034] Figure 7 This is a schematic diagram of the structure of a control device for offline transactions provided in an embodiment of this specification;

[0035] Figure 8 This is a schematic diagram of the structure of a training device for a state prediction model provided in an embodiment of this specification;

[0036] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0037] To make the features and advantages of this specification more obvious and easy to understand, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this specification.

[0038] The following will clearly and thoroughly describe the technical solutions in this specification with reference to the accompanying drawings. In the description of the embodiments of this specification, unless otherwise specified, " / " means or. For example, A / B can mean A or B. "And / or" in the text is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of this specification, "multiple" means two or more than two.

[0039] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0040] In order to improve the control efficiency of offline transactions, an embodiment of this specification provides an offline transaction control method. The execution subject of this offline transaction control method is a control device for offline transactions, and the execution device can specifically be a server. The following are detailed descriptions of each of them. It should be noted that the description order of the following embodiments does not limit the preferred order of the embodiments.

[0041] See also Figure 1 , Figure 11 is a diagram of the system architecture used by the offline transaction control method provided in the embodiments of this specification. The system architecture includes a server 110, a gateway 120, the Internet 130, a terminal device 140, and the like.

[0042] Terminal device 140 includes, but is not limited to, mobile phones, computers, and intelligent voice interaction devices. The embodiments of this specification are applicable to various scenarios, including but not limited to cloud technology and artificial intelligence. Furthermore, it can be a single device or a combination of multiple devices. For example, multiple desktop computers connected via a local area network and working collaboratively using a shared display constitute a terminal device 140. Terminal device 140 can communicate with the internet 130 via wired or wireless means to exchange data.

[0043] Server 110 refers to a computer system that provides certain services to terminal device 140. Compared to ordinary terminal device 140, server 110 has higher requirements in terms of stability, security, and performance. Server 110 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0044] Gateway 120, also known as a gateway or protocol converter, implements network interconnection at the transport layer and is a computer system or device that performs a conversion function. A gateway acts as a translator between two systems using different communication protocols, data formats, languages, or even completely different architectures. Gateways can also provide filtering and security functions. Messages sent from terminal device 140 to server 110 are sent through gateway 120 to the corresponding server 110. Messages sent from server 110 to terminal device 140 are also sent through gateway 120 to the corresponding terminal device 140.

[0045] In the embodiments of this specification, server 110 runs a transaction processing platform pre-installed by developers. This transaction processing platform is used to provide an online operating environment for transactions. The online operating environment can be used to run offline transactions and online transactions. Online transactions are transactions generated by transaction requests initiated in real time by users through terminal devices 140, while offline transactions are non-real-time tasks triggered by the offline transaction platform. Server 110 is equipped and runs a trained state prediction model. This state prediction model is used to predict the operating state of the online operating environment based on its online daily data and adjust the operating state of offline transactions in the online operating environment based on the prediction results.

[0046] In an embodiment of the present specification, the server 110 monitors transactions in the online running environment in real time. If it is detected that an offline transaction is in running state, an online running log within a preset historical time period closest to the current moment is obtained from the log database. The online running log corresponds to the status data generated by each transaction in the online running environment during operation and the usage data of the hardware resources generated during operation.

[0047] Furthermore, transaction status data and device resource data are extracted from the online log data. The transaction status data reflects the execution status of each transaction in the online runtime environment, while the device resource data records the hardware resource usage data of each transaction during runtime. Next, the device resource data is traversed one by one to check for missing values. The missing values ​​are then filled using a pre-set missing compensation algorithm. The padded device resource data is compared with a pre-set threshold. If the data value exceeds the threshold, normalization is performed to scale the data to a uniform range. The transaction status data and device resource data are then integrated according to the data acquisition timestamps and converted into a formatted runtime log.

[0048] The formatted operation log is used as the input of the state prediction model to obtain the prediction result for predicting the impact of offline transactions on the online operation environment. According to the prediction result, the corresponding state intervention strategy is obtained. The state intervention strategy includes a first state intervention strategy and a second state intervention strategy: the first state intervention strategy is used to control the offline transaction to switch from the running state to the stopped state, which is applicable to the situation where the prediction result shows that the offline transaction has a significant negative impact on the online environment; the second state intervention strategy is used to control the offline transaction to maintain the running state, which is applicable to the situation where the prediction result shows that the offline transaction has a small or acceptable impact on the online environment. Based on the selected strategy, the running state of the offline transaction is dynamically adjusted, so as to optimize the execution efficiency of the offline transaction while ensuring the efficient operation of the online transaction in the online operation environment, ensuring the overall stability of the online operation environment and the optimization of hardware resource utilization.

[0049] based on Figure 1 The system architecture diagram shown below will be combined with Figure 2-Figure 4 , an offline transaction control method provided in an embodiment of this specification is introduced in detail.

[0050] See Figure 2 , Figure 2 This is a flow chart of an offline transaction control method provided by an embodiment of this specification. Figure 2 As shown, the method of the embodiment of this specification may include the following steps S101-S104.

[0051] It should be noted that offline transactions and online transactions can be run simultaneously in the online operating environment. Online transactions are transaction requests initiated by users in real time through terminal devices 140, such as order submission, data query, etc., which are characterized by high real-time performance and immediate response to user operations. Therefore, in the online operating environment, it is necessary to prioritize the stability and performance of online transactions to meet the real-time needs of users. In contrast, offline transactions are non-real-time tasks triggered by the offline transaction platform, which are usually executed in a batch manner within a specific time period, such as data statistics, report generation, etc. Their operation will occupy hardware resources and have a certain impact on the online operating environment. Therefore, during execution, it is necessary to ensure that the execution of offline transactions does not interfere with the stability of online transactions. At the same time, through monitoring and optimization measures, the negative impact of offline transactions on the online operating environment can be minimized, thereby ensuring the efficiency and reliability of the online operating environment.

[0052] S101: If the offline transaction in the online running environment is in a running state, obtain the online running log corresponding to the online running environment within a preset historical time period before the current moment.

[0053] In the embodiments of this specification, an online operating environment refers to an environment for processing transactions online in real time. An online operating log records the operations, events, or status information that occur during the execution of each transaction in the online operating environment. An offline transaction refers to a batch processing task that is triggered by an offline platform in the online operating environment rather than actively initiated by a user through terminal device 140. It is typically executed in a batch running mode, that is, processing a large amount of data or performing specific operations in a specific time period.

[0054] It should be noted that the offline transaction in the online running environment may be an offline transaction that is in a running state or an offline transaction to be run that will meet a preset running condition.

[0055] Specifically, the running status of each offline transaction is detected; if it is monitored that there is an offline transaction in the running state, the online running log of the preset historical length is further extracted from the log database of the online running environment, where the preset historical length is the historical period closest to the current moment. The preset historical length can be adjusted according to the actual scenario and is not specifically limited here.

[0056] S102 , extracting transaction status data and device resource data from the online log data, integrating the transaction status data and the device resource data, and obtaining a formatted operation log.

[0057] Specifically, transaction status data and device resource data are extracted from the online log data, wherein the transaction status data reflects the execution status of the transaction, and the device resource data records the usage data of the hardware resources caused by offline transactions and / or online transactions during operation in the online running environment.

[0058] Furthermore, the device resource data is traversed one by one to check whether there are any missing values ​​in each device resource data; if device resource data with missing data is found, the preset missing compensation algorithm will be called to fill the missing part of the device resource data to ensure the integrity and continuity of the device resource data.

[0059] The specific data value of the filled device resource data is obtained and compared with the preset data threshold. If the data value exceeds the threshold, it means that there is an anomaly or the magnitude is too large. At this time, the data value of the device resource data will be normalized, that is, the data value will be scaled proportionally to a uniform range to eliminate dimensional differences and improve the comparability and analyzability of the data.

[0060] The extracted transaction status data and device resource data are consolidated. This involves processing them uniformly according to timestamps through operations such as association, matching, and aggregation to form a structured dataset. Finally, the consolidated data is converted into a formatted operation log, organizing the data according to predefined formats and standards for easy storage, querying, and analysis. The core purpose of this process is to generate structured operation logs through in-depth processing of online operation logs. This provides data support for environmental monitoring, performance optimization, troubleshooting, and post-audit operations, while also improving the readability and usability of online operation logs.

[0061] S103: Input the formatted operation log into a pre-trained state prediction model to obtain a prediction result of the offline transaction on the environmental state of the online operation environment.

[0062] Specifically, formatted operation logs are used as input to the state prediction model. Formatted operation logs refer to structured log data characterized by standardized data and ease of parsing and analysis. The formatted operation logs are then fed into a pre-trained state prediction model. The state prediction model is a machine learning or deep learning model trained on historical data that can predict the impact of offline transactions on the online operating environment based on the input data. The state prediction model analyzes key features in the formatted operation log data (such as resource utilization, transaction execution time, and error rate) to output predictions about the environmental state of the online operating environment as a result of offline transactions. The core purpose of this process is to use formatted operation log predictions to identify in advance whether the impact of offline transactions on the online operating environment will interfere with the operational stability and performance of online transactions.

[0063] S104: Obtain a state intervention strategy corresponding to the prediction result, and adjust the running state of the offline transaction based on the state intervention strategy.

[0064] Specifically, according to the prediction results output by the state prediction model, a corresponding state intervention strategy is obtained, wherein the state intervention strategy refers to an adjustment plan formulated for the impact of offline transactions on the online operating environment, with the purpose of ensuring the stability of online transactions and the reasonable allocation of hardware resources; the state intervention strategy includes a first state intervention strategy and a second state intervention strategy. The first state intervention strategy is used to control the offline transaction to switch from the running state to the stopped state, and is applicable to situations where the prediction results show that the offline transaction has a significant negative impact on the online environment. The second state intervention strategy is used to control the offline transaction to remain in the running state, and is applicable to situations where the prediction results show that the offline transaction has a small or acceptable impact on the online environment.

[0065] Based on the selected status intervention strategy, the running status of offline transactions is dynamically adjusted, such as pausing or continuing the execution of offline transactions. This ensures the efficient operation of online transactions while optimizing the execution efficiency of offline transactions, thereby ensuring the overall stability of the online operating environment and the optimization of hardware resource utilization.

[0066] As can be seen from the above, by obtaining online operation logs within a preset historical period, we can fully understand the operating status and historical trends of the online environment, providing a data basis for subsequent analysis. Secondly, by extracting transaction status data and device resource data from the logs and integrating them, we can generate formatted operation logs to ensure the structured and analyzable nature of the data. Next, the formatted logs are input into the pre-trained state prediction model, which can accurately predict the impact of offline transactions on the state of the online environment, thereby identifying potential problems in advance. Finally, based on the prediction results, the corresponding state intervention strategy is obtained, which can dynamically adjust the operating status of offline transactions, avoid transaction conflicts, and ensure the stable operation of the online operating environment. This process not only realizes the intelligent management of offline transactions, but also significantly reduces the cost of manual intervention and improves the environmental stability of the online operating environment.

[0067] Since the amount of data to be acquired is large, if the data is disorganized during subsequent data analysis, it will prolong the data analysis time. Figure 3 , Figure 3 This is a flow chart of an offline transaction control method provided by an embodiment of this specification. Figure 3 As shown, the method of the embodiment of this specification may include the following steps S201-S205.

[0068] S201: Obtain a first timestamp corresponding to transaction status data and a second timestamp corresponding to device resource data.

[0069] S202 , arranging and integrating the transaction status data and the device resource data based on the time sequence of the first timestamp and the second timestamp to generate a formatted operation log.

[0070] Specifically, in S201-S202, the time when the transaction status data is acquired is recorded as the first timestamp, and the time when the device resource data is acquired is recorded as the second timestamp. The transaction status data and the device resource data are arranged and integrated according to the time sequence of the first timestamp and the second timestamp to generate formatted operation data.

[0071] For example, in the transaction status data, the transaction with transaction ID 001 starts executing at 2023-10-01 10:00 and completes at 2023-10-01 10:05; at the same time, in the device resource data, the processor utilization rate of the device with device ID A001 is 80% and the memory utilization rate is 60% at 2023-10-01 10:02, and the processor utilization rate is 85% and the memory utilization rate is 65% at 2023-10-01 10:04. Based on the timestamps of this data (2023-10-01 10:00, 2023-10-01 10:02, 2023-10-01 10:04, 2023-10-01 10:05), the transaction status data and device resource data are arranged and integrated in chronological order. The resulting formatted run log is as follows: 2023-10-01 10:00 - Transaction 001 starts; 2023-10-01 10:02 - Device A001 processor utilization is 80%, and memory utilization is 60%; 2023-10-01 10:04 - Device A001 processor utilization is 85%, and memory utilization is 65%; 2023-10-01 10:05 - Transaction 001 completes.

[0072] S203: traverse the device resource data. If there is data missing in the device resource data, call a preset missing compensation algorithm to fill the data in the device resource data.

[0073] Specifically, during the device resource data processing process, the device resource data is first traversed to check whether there is any missing data. For example, the device resource data records that the processor utilization rate of device A001 at 2023-10-01 10:02 is 80%, but the data at 2023-10-01 10:03 is missing. At this time, a preset missing compensation algorithm is called, for example, based on the time series interpolation method. Based on the data of the previous and next time points, such as 80% at 2023-10-01 10:02 and 85% at 2023-10-01 10:04, the processor utilization rate at 2023-10-01 10:03 is calculated to be 82.5%, thereby filling the missing data and ensuring the integrity of the device resource data.

[0074] S204, obtaining the data value of the device resource data after data padding.

[0075] S205: If the data value is greater than a preset data threshold, perform data normalization processing on the device resource data.

[0076] Specifically, in S204-S205, after data filling is completed, the specific value of the filled device resource data will be obtained and the filled device resource data value will be compared with the preset data threshold to perform data normalization on the device resource data to ensure that each data point of the device resource data is within the preset range, wherein the data value of the device resource data can specifically be the processor usage rate.

[0077] For example, the data threshold is 80%, and the data value of the device resource data of device A001 is 70%, which is normalized to 0. The data value of the device resource data of device A002 is 90%, which is normalized to 1.

[0078] As can be seen above, data is arranged and integrated based on timestamps to generate structured operation logs, facilitating subsequent analysis and traceability. Secondly, data traversal and missing value compensation algorithms are used to fill missing values ​​in device resource data, avoiding errors caused by incomplete data. Finally, through data cropping and normalization, device resource data is controlled within a preset range, standardizing the data scale and improving data processability and interpretability. This not only improves data quality but also provides reliable data support for monitoring the operating environment's operating status, troubleshooting, and performance optimization, significantly improving the stability of the operating environment and the accuracy of decision-making.

[0079] Since online transactions are real-time transactions initiated by users, if the prediction result shows that offline transactions will affect the online operating environment and thus the operating status of online transactions, timely intervention strategies need to be adopted to ensure the stability of the online operating environment. Figure 4, Figure 4 This is a flow chart of an offline transaction control method provided by an embodiment of this specification. Figure 4 As shown, the method of the embodiment of this specification may include the following steps S301-S303.

[0080] S301: Obtain a state intervention strategy corresponding to the prediction result, and adjust the running state of the offline transaction based on the state intervention strategy.

[0081] Specifically, the specific execution process of S301 can be referred to the above S103, which will not be repeated here.

[0082] S302: If the state intervention strategy is the first state intervention strategy, control the offline transaction to switch from the running state to the stopped state.

[0083] S303: If the state intervention strategy is the second state intervention strategy, control the offline transaction to maintain the running state.

[0084] Specifically, in S302-S303, the first state intervention strategy is used to control the offline transaction to switch from the running state to the stopped state, which is applicable to situations where the prediction results show that the offline transaction has a significant negative impact on the online environment. The second state intervention strategy is used to control the offline transaction to remain in the running state, which is applicable to situations where the prediction results show that the offline transaction has a small or acceptable impact on the online environment.

[0085] For example, if offline transaction A is running to process a batch of data and the prediction result indicates an abnormality in the online operating environment, the first state intervention strategy is triggered. At this time, offline transaction A is immediately switched from the running state to the stopped state to free up resources and avoid resource overload. Conversely, if the prediction result indicates that the online operating environment is normal, the second state intervention strategy is adopted to keep offline transaction A running and ensure the successful completion of the data processing task.

[0086] From the above, it can be seen that by calling the first state intervention strategy or the second state intervention strategy according to the prediction results of the state prediction model, the state of the offline transaction can be dynamically adjusted according to the actual operating environment and resource conditions, thereby optimizing resource utilization and improving the operating stability of the online operating environment.

[0087] based on Figure 1 The system architecture diagram shown below will be combined with Figure 5-Figure 6 , a state prediction model training method provided in an embodiment of this specification is introduced in detail.

[0088] See Figure 5 , Figure 5 This is a flow chart of a state prediction model training method provided by an embodiment of this specification. Figure 5As shown, the method of the embodiment of this specification may include the following steps S401-S404.

[0089] S401: Obtain preset training samples and test samples.

[0090] Specifically, we first extract transaction status data and device resource data from historical operation logs. Transaction status data refers to various status information of transactions during execution, such as start time, end time, and execution results. Device resource data refers to the usage of hardware resources during operation, such as processors, memory, and disks. Subsequently, we perform data cleaning and preprocessing on these data, including noise removal (i.e., deleting outliers or invalid data), filling missing values ​​(i.e., reasonably filling blanks in the data), and normalization (i.e., scaling the data to a uniform range, such as between 0 and 1), to ensure data quality.

[0091] The dataset is divided into training samples and test samples. The training samples are used for state prediction model training, and the test samples are used for state prediction model performance evaluation. In this manual, cross-validation (i.e., dividing the dataset into multiple subsets, which are used as training sets and test sets in turn) or chronological partitioning (i.e., dividing the data in chronological order to prevent future data from leaking into the training set) is usually adopted to ensure that the distribution of training samples and test samples is consistent and there is no data leakage. At the same time, key features are extracted from the original data according to business needs, such as transaction execution time, resource utilization, concurrency, etc., and feature selection (i.e., selecting the features most useful for model prediction) or dimensionality reduction (i.e., reducing the number of features and reducing computational complexity) is performed to improve the training efficiency of the state prediction model.

[0092] S402: Input the training sample into the state prediction model to obtain the predicted operating state, and call the mean square error loss function to calculate the state error between the predicted operating state and the actual operating state.

[0093] Specifically, the feature data of the training samples are input into the state prediction model. The state prediction model can be a model based on machine learning or a deep learning model. For example, LSTM (Long Short-Term Memory) is a deep learning model suitable for time series data, which can capture long-term dependencies in the data. Transformer is a deep learning model based on the self-attention mechanism, which is suitable for processing complex sequence data. The model generates a predicted operating state based on the input features, such as resource occupancy rate, transaction success rate, etc. Then, the mean square error loss function is called to calculate the error between the predicted operating state and the actual operating state. The mean square error loss function is a commonly used regression model evaluation indicator and will not be described in detail here.

[0094] S403 , calling the back propagation algorithm to adjust the model parameters of the state prediction model based on the state error, so that the mean square error loss function reaches a convergence state.

[0095] Specifically, the backpropagation algorithm is an optimization algorithm used to train neural networks. It propagates the error back layer by layer from the output layer through the chain rule to calculate the gradient of the loss function with respect to the parameters of the state prediction model. Subsequently, optimization algorithms such as SGD (Stochastic Gradient Descent) and Adam (Adaptive Moment Estimation) are used to update the model parameters based on the gradient. Stochastic gradient descent is a method of iteratively updating parameters to minimize the loss function, and Adam is an optimization algorithm that combines momentum and adaptive learning rate to accelerate model convergence.

[0096] The above process is repeated until the mean square error loss function reaches a convergence state (that is, the loss value tends to be stable or reaches a preset threshold), thereby completing the training process of the state prediction model.

[0097] S404: Testing the state prediction model based on the test sample to determine the training state of the state prediction model.

[0098] Specifically, the test sample is input into the trained state prediction model to generate a predicted running state. The error between the predicted result and the actual state is calculated, and evaluation indicators such as mean square error, mean absolute error, and R² are used to quantify the performance of the state prediction model. The mean square error and mean absolute error are used to measure the deviation between the predicted value and the actual value, and R² is used to measure the model's ability to interpret the data. If the model performance does not meet expectations, return to the above S401 to readjust the sample division, feature engineering, or model structure. If the performance meets the standards, save the state prediction model and deploy it to the actual environment to provide support for offline transactions to predict the state of the online running environment.

[0099] As can be seen above, through high-quality sample data preparation and feature engineering, the state prediction model is able to capture key operational patterns, providing a reliable foundation for subsequent predictions. Secondly, through iterative training of the loss function and optimization algorithm, the state prediction model is able to gradually reduce prediction errors and improve prediction accuracy. Finally, performance evaluation based on test samples ensures the generalization and reliability of the state prediction model, enabling it to adapt to complex changes in the real environment. This process not only enables accurate predictions of the impact of offline transactions but also provides a scientific basis for resource scheduling, thereby improving overall operational efficiency and stability and reducing the risk of resource conflicts and performance degradation.

[0100] To further improve the prediction accuracy of the state prediction model, see Figure 6 , Figure 6 This is a flow chart of a state prediction model training method provided by an embodiment of this specification. Figure 6 As shown, the method of the embodiment of this specification may include the following steps S501-S503.

[0101] S501: Input a test sample into a state prediction model to obtain a predicted operating state. The test sample is composed of a plurality of different transaction state data and device resource data.

[0102] Specifically, the test sample is input into the state prediction model. The test sample consists of multiple different transaction state data and device resource data. The transaction state data refers to various state information of the transaction during the execution process, such as start time, end time, execution result, etc. The device resource data refers to the usage of hardware resources such as processor, memory, disk, etc. in the online operating environment during operation. The state prediction model is an algorithm model that predicts future operating status by learning the patterns of historical data. It can be a model based on machine learning or deep learning. The predicted operating status is generated according to the characteristic data of the test sample, such as resource occupancy rate, transaction success rate, etc., to provide a data basis for subsequent performance evaluation.

[0103] S502: Determine the state prediction accuracy of the state prediction model based on the actual operating state and the predicted operating state.

[0104] Specifically, the state prediction accuracy of the state prediction model is determined based on the actual operating state and the predicted operating state. The actual operating state refers to the actual transaction and device resource state recorded in the test sample, and the predicted operating state refers to the prediction result generated by the state prediction model based on the test sample. The state prediction accuracy is the core indicator for measuring the prediction performance of the state prediction model, which is usually determined by calculating the degree of match between the predicted value and the true value, such as using the classification accuracy (for classification tasks) or error rate (for regression tasks). By comparing the actual state and the predicted state, the prediction ability of the state prediction model is quantified, providing an objective basis for the performance evaluation of the state prediction model.

[0105] For example, the state prediction model predicts that the execution success rate of offline transaction A is 95%, and the resource usage rate caused by offline transaction B is 80%. However, the actual operation status shows that the execution success rate of offline transaction A is 96%, and the resource usage rate caused by offline transaction B is 78%. The state prediction accuracy is determined by calculating the error between the predicted value and the actual value, such as using the mean square error or absolute error. For example, the calculated accuracy is 92%, indicating that the prediction result of the state prediction model matches the actual state to a high degree.

[0106] S503: If the state prediction accuracy is greater than or equal to a preset accuracy threshold, the training state is determined to be a training completion state.

[0107] Specifically, the state prediction accuracy is compared with the preset accuracy threshold. The accuracy threshold is the minimum acceptable standard set according to business needs and state prediction model performance requirements. If the state prediction accuracy is greater than or equal to the preset accuracy threshold, the training state is determined to be the training completion state, indicating that the state prediction model has achieved the expected performance and can be deployed to the actual environment. If the state prediction accuracy is lower than the threshold, it is necessary to return to the training stage of the state prediction model, readjust the state prediction model parameters, optimize feature engineering, or increase training data until the performance requirements are met.

[0108] For example, the preset accuracy threshold is 90%. Since the state prediction model's state prediction accuracy (92%) exceeds the threshold, the training status is determined to be complete, indicating that the state prediction model meets the performance requirements and can be deployed in the actual environment. If the accuracy is lower than 90%, the state prediction model needs to be retrained or the data needs to be optimized until the performance requirements are met. This process ensures that the prediction capabilities of the state prediction model meet actual needs.

[0109] As can be seen above, by inputting test samples into the state prediction model and generating a predicted operating state, the state prediction model's ability to handle complex transactions and resource data can be verified, providing a data foundation for performance evaluation. Secondly, by comparing the actual operating state with the predicted operating state, the state prediction model's prediction accuracy can be quantified, objectively measuring its predictive capabilities. Finally, by comparing it with the preset accuracy threshold, it can be determined whether the state prediction model has achieved the expected performance, thereby deciding whether to deploy it or further optimize it. This process not only ensures the prediction accuracy and stability of the state prediction model, but also provides a scientific basis for resource scheduling and optimization in real-world environments, reduces the risk of transaction conflicts caused by prediction errors, and improves overall operational efficiency and intelligence.

[0110] based on Figure 1 The system architecture will be combined with Figure 7 , the control device for offline transactions provided by the embodiment of this specification is introduced in detail. It should be noted that, Figure 7 The control device of the offline transaction is used to execute the Figure 2-Figure 4 For the convenience of explanation, only the part related to the embodiment of this specification is shown. For the specific technical details not disclosed, please refer to this specification. Figure 2-Figure 4 In the embodiment shown, the offline transaction control device 600 may include a log acquisition unit 601, a data processing unit 602, a result prediction unit 603, and a state adjustment unit 604, specifically as follows:

[0111] Log acquisition unit 601: for acquiring online operation logs corresponding to the online operation environment within a preset historical period before the current moment if the offline transaction in the online operation environment is in the running state;

[0112] Data processing unit 602: used to extract transaction status data and device resource data from online log data, integrate the transaction status data and device resource data, and obtain a formatted operation log;

[0113] Result prediction unit 603: used to input the formatted operation log into the pre-trained state prediction model to obtain the prediction result of the offline transaction on the environmental state of the online operation environment;

[0114] The state adjustment unit 604 is configured to obtain a state intervention strategy corresponding to the prediction result, and adjust the running state of the offline transaction based on the state intervention strategy.

[0115] In some embodiments, the data processing unit 602 further includes a timestamp obtaining unit and a log generating unit.

[0116] A timestamp acquisition unit, configured to acquire a first timestamp corresponding to the transaction status data and a second timestamp corresponding to the device resource data;

[0117] The log generating unit is configured to arrange and integrate the transaction status data and the device resource data based on the time sequence of the first timestamp and the second timestamp to generate a formatted operation log.

[0118] In some embodiments, the data processing unit 602 further includes a data traversal unit.

[0119] The data traversal unit is used to traverse the device resource data. If there is data missing in the device resource data, a preset missing compensation algorithm is called to fill the data in the device resource data.

[0120] In some embodiments, the data processing unit 602 further includes a data value acquisition unit and a value comparison unit.

[0121] A data value acquisition unit, used to acquire the data value of the device resource data after data padding;

[0122] The numerical comparison unit is used to perform data normalization processing on the device resource data if the data value is greater than a preset data threshold.

[0123] In some embodiments, the data processing unit 602 further includes a first determination unit and a second determination unit.

[0124] a first determining unit, configured to control the offline transaction to switch from a running state to a stopped state if the state intervention strategy is a first state intervention strategy;

[0125] The second determination unit is configured to control the offline transaction to maintain a running state if the state intervention strategy is the second state intervention strategy.

[0126] In the embodiments of this specification, by obtaining online operation logs within a preset historical period, it is possible to fully understand the operating status and historical trends of the online environment, providing a data basis for subsequent analysis. Secondly, by extracting transaction status data and device resource data from the log and integrating them, a formatted operation log can be generated to ensure the structured and analyzable nature of the data. Next, the formatted log is input into a pre-trained state prediction model, which can accurately predict the impact of offline transactions on the state of the online environment, thereby identifying potential problems in advance. Finally, based on the prediction results, the corresponding state intervention strategy is obtained, which can dynamically adjust the operating status of offline transactions, avoid transaction conflicts, and ensure the stable operation of the online operating environment. This process not only realizes the intelligent management of offline transactions, but also significantly reduces the cost of manual intervention and improves the environmental stability of the online operating environment.

[0127] based on Figure 1 The system architecture will be combined with Figure 8 , the training device of the state prediction model provided in the embodiment of this specification is introduced in detail. It should be noted that, Figure 8 The training device of the state prediction model in this specification is used to execute Figure 5-Figure 6 For the convenience of explanation, only the part related to the embodiment of this specification is shown. For the specific technical details not disclosed, please refer to this specification. Figure 5-Figure 6 In the embodiment shown, the state prediction model training device 700 may include a sample acquisition unit 701, a sample processing unit 702, a parameter adjustment unit 703, and a state determination unit 704, specifically as follows:

[0128] The sample acquisition unit 701 is used to acquire preset training samples and test samples. The training samples are actual operating states corresponding to online operating environments under different transaction state data and device resource data. The test samples are composed of multiple different transaction state data and device resource data.

[0129] The sample processing unit 702 is used to input the training sample into the state prediction model to obtain the predicted operating state, and call the mean square error loss function to calculate the state error between the predicted operating state and the actual operating state;

[0130] A parameter adjustment unit 703 is configured to call a back propagation algorithm to adjust the model parameters of the state prediction model based on the state error so that the mean square error loss function reaches a convergence state;

[0131] The state determination unit 704 is configured to test the state prediction model based on the test sample to determine the training state of the state prediction model.

[0132] In some embodiments, the state determination unit 704 further includes a predicted state acquisition unit, a prediction accuracy acquisition unit, and a state determination unit.

[0133] A predicted state acquisition unit is used to input a test sample into a state prediction model to obtain a predicted operating state, wherein the test sample is composed of a plurality of different transaction state data and device resource data;

[0134] A prediction accuracy acquisition unit, configured to determine a state prediction accuracy of a state prediction model based on the actual operating state and the predicted operating state;

[0135] The state determination unit is configured to determine that the training state is a training completion state if the state prediction accuracy is greater than or equal to a preset accuracy threshold.

[0136] In addition, the offline transaction control device and an offline transaction control method embodiment provided in the above embodiment, as well as the state prediction model training device and a state prediction model training method, belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0137] The serial numbers of the embodiments in this specification are for descriptive purposes only and do not represent the merits of the embodiments. In some cases, the actions or steps recited in the claims may be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0138] See Figure 9 , is a schematic diagram of the structure of an electronic device provided in the embodiment of this specification. Figure 9 As shown, the electronic device 800 includes a processor 801 and a memory 802. The processor 801 is electrically connected to the memory 802.

[0139] Processor 801 is the control center of electronic device 800 and may include one or more processing cores. Using various interfaces and circuits, processor 801 connects the various components of the electronic device. By running or invoking computer programs stored in memory 802, as well as accessing data stored in memory 802, processor 801 executes various functions of the electronic device and processes data, thereby providing overall control over the electronic device. Optionally, processor 801 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field programmable gate array (FPGA), or a programmable logic array (PLA). Processor 801 may integrate one or a combination of a processor, a graphics processing unit (GPU), and a modem. The processor primarily handles the operating system, user interfaces, and applications; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may also be implemented independently of the processor 801 and implemented via a separate communications chip.

[0140] The memory 802 can be used to store software programs and modules. The processor 801 executes various functional applications and controls offline transactions by running the computer programs and modules stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area. The program storage area may store an operating system, computer programs required for at least one function, etc.; the data storage area may store data generated based on the use of the electronic device.

[0141] In addition, the memory 802 may include a high-speed random access memory and a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0142] In the embodiment of this specification, the processor 801 in the electronic device 800 loads instructions corresponding to one or more computer program processes into the memory 802 according to the following steps, and the processor 801 runs the computer program stored in the memory 802 to implement various functions as follows:

[0143] If the offline transaction in the online running environment is in a running state, obtain the online running log corresponding to the online running environment within the preset historical length before the current moment; extract the transaction status data and device resource data from the online log data, integrate the transaction status data and device resource data to obtain a formatted running log; input the formatted running log into a pre-trained state prediction model to obtain the prediction result of the offline transaction on the environmental state of the online running environment; obtain the state intervention strategy corresponding to the prediction result, and adjust the running state of the offline transaction based on the state intervention strategy.

[0144] Optionally, the processor 801 performs data integration on the transaction status data and the device resource data to obtain a formatted operation log, specifically performing the following steps: obtaining a first timestamp corresponding to the transaction status data and a second timestamp corresponding to the device resource data; arranging and integrating the transaction status data and the device resource data based on the time sequence of the first timestamp and the second timestamp to generate a formatted operation log.

[0145] Optionally, after executing to obtain the first timestamp corresponding to the transaction status data and the second timestamp corresponding to the device resource data, the processor 801 specifically performs: data traversal of the device resource data, and if there is data missing in the device resource data, calling a preset missing compensation algorithm to fill the data of the device resource data.

[0146] Optionally, after the processor 801 executes the preset missing compensation algorithm to fill the device resource data and obtains the target device resource data, it specifically performs the following steps: obtaining the data value of the device resource data after data filling; if the data value is greater than the preset data threshold, performing data normalization on the device resource data.

[0147] Optionally, the processor 801 executes the state intervention strategy corresponding to the prediction result and adjusts the running state of the offline transaction based on the state intervention strategy. Specifically, if the state intervention strategy is the first state intervention strategy, the offline transaction is controlled to switch from the running state to the stopped state; if the state intervention strategy is the second state intervention strategy, the offline transaction is controlled to maintain the running state.

[0148] In the embodiments of this specification, by obtaining online operation logs within a preset historical period, it is possible to fully understand the operating status and historical trends of the online environment, providing a data basis for subsequent analysis. Secondly, by extracting transaction status data and device resource data from the log and integrating them, a formatted operation log can be generated to ensure the structured and analyzable nature of the data. Then, the formatted log is input into a pre-trained state prediction model, which can accurately predict the impact of offline transactions on the state of the online environment, thereby identifying potential problems in advance. Finally, based on the prediction results, the corresponding state intervention strategy is obtained, which can dynamically adjust the operating status of offline transactions, avoid transaction conflicts, and ensure the stable operation of the online operating environment. This process not only realizes the intelligent management of offline transactions, but also significantly reduces the cost of manual intervention and improves the environmental stability of the online operating environment.

[0149] In the embodiment of this specification, the processor 801 in the electronic device 800 also loads instructions corresponding to one or more computer program processes into the memory 802 according to the following steps, and the processor 801 runs the computer program stored in the memory 802 to implement various functions as follows:

[0150] Obtain preset training samples and test samples. The training samples are the actual operating states corresponding to the online operating environment under different transaction status data and device resource data, and the test samples are composed of multiple different transaction status data and device resource data. Input the training samples into the state prediction model to obtain the predicted operating state, and call the mean square error loss function to calculate the state error between the predicted operating state and the actual operating state. Call the back propagation algorithm to adjust the model parameters of the state prediction model based on the state error so that the mean square error loss function reaches a convergence state. Test the state prediction model based on the test samples to determine the training state of the state prediction model.

[0151] Optionally, the processor 801 performs a test on the state prediction model based on the test sample to determine the training status of the state prediction model, specifically performing: inputting the test sample into the state prediction model to obtain a predicted operating state, the test sample consisting of multiple different transaction state data and device resource data; based on the actual operating state and the predicted operating state, determining the state prediction accuracy of the state prediction model; if the state prediction accuracy is greater than or equal to a preset accuracy threshold, determining that the training state is a training completion state.

[0152] In addition, the device provided in the embodiments of this specification may specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute an offline transaction control method and a state prediction model training method provided in the above embodiments.

[0153] An embodiment of this specification also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement an offline transaction control method and state prediction model training method provided in the above-mentioned embodiment.

[0154] This specification also provides a computer program product, which stores at least one instruction, and the at least one instruction is loaded and executed by the processor as described above. Figure 2-Figure 4 The offline transaction control method of the embodiment shown and the above Figure 5-Figure 6 The state prediction model training method of the embodiment shown, the specific execution process can be found in Figure 2-Figure 6 The detailed description of the illustrated embodiment will not be repeated here.

[0155] Among them, the devices, computer-readable storage media, computer program products or chips provided in the embodiments of this specification are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above and will not be repeated here.

[0156] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0157] In the embodiments provided in this specification, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between the related devices shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0158] The above content is merely a specific embodiment of this specification, but the scope of protection of this specification is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this specification should be included in the scope of protection of this specification. Therefore, the scope of protection of this specification should be based on the scope of protection of the claims.

Claims

1. An offline transaction control method, the method comprising: If the offline transaction in the online running environment is in a running state, then obtain the online log data corresponding to the online running environment within a preset historical period before the current moment; Extracting transaction status data and device resource data from the online log data, and integrating the transaction status data and the device resource data to obtain a formatted operation log; Inputting the formatted operation log into a pre-trained state prediction model to obtain a prediction result of the offline transaction on the environmental state of the online operation environment; Obtaining a state intervention strategy corresponding to the prediction result; If the state intervention strategy is the first state intervention strategy, controlling the offline transaction to switch from the running state to the stopped state; If the state intervention strategy is the second state intervention strategy, the offline transaction is controlled to maintain the running state.

2. The method according to claim 1, wherein the step of integrating the transaction status data and the device resource data to obtain a formatted operation log comprises: Obtaining a first timestamp corresponding to the transaction status data and a second timestamp corresponding to the device resource data; The transaction status data and the device resource data are arranged and integrated based on the time sequence of the first timestamp and the second timestamp to generate the formatted operation log.

3. The method according to claim 2, after obtaining the first timestamp corresponding to the transaction status data and the second timestamp corresponding to the device resource data, the method further comprises: The device resource data is traversed, and if there is data missing in the device resource data, a preset missing compensation algorithm is called to fill the data in the device resource data.

4. The method according to claim 3, wherein after the preset missing compensation algorithm is called to fill the device resource data to obtain the target device resource data, the method further comprises: Obtaining data values ​​of the device resource data after data padding; If the data value is greater than a preset data threshold, data normalization processing is performed on the device resource data.

5. A control device for offline transactions, comprising: Log acquisition unit: used for acquiring online log data corresponding to the online running environment within a preset historical period before the current moment if the offline transaction in the online running environment is in a running state; Data processing unit: used for extracting transaction status data and device resource data from the online log data, integrating the transaction status data and device resource data, and obtaining a formatted operation log; Result prediction unit: used for inputting the formatted operation log into a pre-trained state prediction model to obtain a prediction result of the offline transaction on the environmental state of the online operation environment; State adjustment unit: used to obtain the state intervention strategy corresponding to the prediction result. If the state intervention strategy is the first state intervention strategy, the offline transaction is controlled to switch from the running state to the stopped state; if the state intervention strategy is the second state intervention strategy, the offline transaction is controlled to maintain the running state.

6. An electronic device, comprising: a memory for storing executable program code; A processor is configured to call and run the executable program code from the memory, so that the electronic device executes the method according to any one of claims 1 to 4. 7 . A computer-readable storage medium storing a computer program, wherein when the computer program is executed, the method according to claim 1 is implemented.

8. A computer program product having at least one instruction stored thereon, wherein when the at least one instruction is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images