A traffic data processing method and device

Abstract

The application discloses a traffic data processing method and device, and relates to the field of financial technology, and the method comprises the following steps: determining the predicted total traffic cost of multiple transactions in a target time period according to historical traffic data of the multiple transactions; determining whether to update the traffic parameters of the multiple transactions according to the predicted total traffic cost and system traffic resource data; and updating the traffic parameters of the multiple transactions according to the priorities of the multiple transactions in the case that the predicted total traffic cost is greater than the system traffic resource data. The method solves the problem that different types of transactions in different time periods have different transaction traffic demands in the prior art, and the current traffic limiting is only performed in the dimension of the types of transactions, so that the system throughput is low, and system resources cannot be effectively utilized.

Description

Technical Field

[0001] This application relates to the field of financial technology, and more specifically, to a method and apparatus for processing traffic data. Background Technology

[0002] For single containers, rate limiting schemes significantly impact system availability and stability. While most current rate limiting schemes are configurable, they only limit rates based on transaction type, neglecting the impact of dynamic factors such as time on transaction traffic. This solution proposes a transaction cost function based on actual transaction scenarios and dynamically adjusts the system's rate limiting parameters by minimizing this cost function using a Deep Neural Network (DNN) algorithm, thereby achieving optimal system concurrency and throughput.

[0003] While existing systems employ configurable rate limiting schemes, they only limit transactions by type, neglecting the impact of dynamic factors such as time on transaction volume. Recently, our bank has detected certain software sending a large number of invalid requests to the financial institution daily, placing additional pressure on the institution's mainframe system, and the volume of invalid transactions continues to rise. However, simply rate limiting certain transactions using traditional methods would disrupt normal daily deductions. Therefore, there is a need for dynamically changing rate limiting values ​​over time.

[0004] Different types of transactions at different times in related technologies have different transaction traffic requirements. However, current rate limiting only limits the flow based on the type of transaction, resulting in low system throughput and ineffective utilization of system resources. No effective solution has yet been proposed. Summary of the Invention

[0005] The main purpose of this application is to provide a traffic data processing method and apparatus to solve the problem that different types of transactions at different times have different transaction traffic requirements in related technologies. However, current traffic limiting only limits the traffic based on the type of transaction, resulting in low system throughput and ineffective utilization of system resources.

[0006] To achieve the above objectives, according to one aspect of this application, a traffic data processing method is provided. The method includes: determining the predicted total traffic cost of the multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; determining whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the traffic resource data of the system for processing the multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; and updating the traffic parameters of the multiple transactions according to their priority if the predicted total traffic cost is greater than the system traffic resource data.

[0007] Optionally, determining whether to update the traffic parameters of the plurality of transactions based on the predicted total traffic cost and the system traffic resource data further includes: if the predicted total traffic cost is less than or equal to the system traffic resource data, the traffic parameters of the plurality of transactions remain unchanged; if the predicted total traffic cost is greater than the system traffic resource data, updating the traffic parameters of the plurality of transactions according to their priority includes: determining a first difference between the sum of the traffic costs of the plurality of transactions and the system traffic resource data; and updating the traffic parameters of the plurality of transactions according to their priority based on the first difference.

[0008] Optionally, before updating the traffic parameters of the multiple transactions, the method further includes: determining the initial traffic resource data allocated to each transaction based on the system traffic resource data, the predicted total traffic cost, and the predicted cost data of each transaction; and using the initial traffic resource data as the traffic parameter of each transaction.

[0009] Optionally, the traffic parameters include the initial traffic resource data. Based on the first difference, the traffic parameters of the multiple transactions are updated according to their priorities, including: comparing the first predicted cost data of the lowest-priority first transaction among the multiple transactions with the sum of a preset transaction traffic threshold and the first difference; if the first predicted cost data is greater than the sum of the preset transaction traffic threshold and the first difference, updating the first traffic parameter of the first transaction to the first predicted cost data minus the first difference, and completing the update of the traffic parameters of the multiple transactions, wherein the preset transaction traffic threshold is the lowest traffic threshold for the multiple transactions to conduct the transaction; in the first... If the predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to a second difference, wherein the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; if the second difference is greater than zero, the second predicted cost data of the second transaction among the plurality of transactions is compared with the sum of the preset transaction flow threshold and the second difference, wherein the second transaction is the transaction with the lowest priority among the plurality of transactions excluding the first transaction; if the second difference is less than or equal to zero, the flow parameters of the plurality of transactions are updated.

[0010] Optionally, before determining the predicted total cost of traffic for the multiple transactions within a target time period based on historical traffic data of multiple transactions, the method further includes: acquiring historical transaction data of the multiple transactions within a unit of historical time; determining a training sample set based on the transaction time parameters of each transaction within the unit of historical time and the historical transaction volume; and training the prediction model based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction within the target time period.

[0011] Optionally, determining the predicted total traffic cost for a target time period based on historical traffic data from multiple transactions includes: obtaining the historical traffic data to determine the average time spent on the target transaction, wherein the target transaction is any one of the multiple transactions; determining the target predicted cost data for the target transaction in the target time period based on the average time spent on the target transaction and the prediction model; and determining the predicted total traffic cost for the multiple transactions in the target time period based on the target predicted cost data obtained by traversing the multiple transactions as the target transaction.

[0012] Optionally, based on the average time consumption of the target transaction and the prediction model, the target predicted cost data of the target transaction in the target time period is determined. The method further includes: inputting a preset time point of the target transaction in the target time period into the prediction model to obtain the predicted transaction volume of the target transaction in the target time period; and determining the target predicted cost data of the target transaction based on the average time consumption of the target transaction and the predicted transaction volume.

[0013] To achieve the above objectives, according to another aspect of this application, a traffic data processing apparatus is provided. The apparatus includes: a first determining module, configured to determine the predicted total traffic cost of the multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; a second determining module, configured to determine whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the traffic resource data of the system for processing the multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; and an updating module, configured to update the traffic parameters of the multiple transactions according to the priority of the multiple transactions if the predicted total traffic cost is greater than the system traffic resource data.

[0014] According to another aspect of this application, a computer-readable storage medium is also provided, the storage medium being used to store a program, wherein the program executes the traffic data processing method described in any one of the foregoing.

[0015] According to another aspect of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the traffic data processing method described in any one of the above.

[0016] This application employs the following steps: Based on historical traffic data from multiple transactions, the predicted total traffic cost for multiple transactions within a target time period is determined, where the predicted total traffic cost is the sum of the predicted cost data for each of the multiple transactions; based on the predicted total traffic cost and system traffic resource data, it is determined whether to update the traffic parameters for the multiple transactions, where the system traffic resource data refers to the traffic resource data used by the system to process the multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; if the predicted total traffic cost is greater than the system traffic resource data, the traffic parameters for the multiple transactions are updated according to their priority. By comparing the predicted traffic cost data for multiple transactions within the target time period with the system traffic resource data, it is determined whether to implement traffic limiting for transactions within the target time period. Then, by adjusting the traffic parameters, traffic limiting is implemented for one or more of the multiple transactions. This achieves the goal of traffic limiting transactions based on their traffic demand at different times, thus improving the utilization rate of system traffic resources. Furthermore, it solves the problem in related technologies where different types of transactions at different times have different traffic demands, but current traffic limiting only considers the type of transaction, resulting in low system throughput and ineffective utilization of system resources. Attached Figure Description

[0017] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a flowchart of a traffic data processing method provided according to an embodiment of this application;

[0019] Figure 2 This is a schematic diagram of the overall architecture of a precise rate limiting method based on DNN dynamic programming provided according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of a traffic data processing device according to an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] For ease of description, the following explains some of the nouns or terms used in the embodiments of this application:

[0026] Deep learning: a type of machine learning that uses neural networks to learn the inherent patterns in sample data.

[0027] Rate limiting: To prevent excessive non-core transaction traffic from impacting the system and to protect core transactions, the trading system limits the processing of non-core transactions based on dimensions such as concurrency or TPS (throughput).

[0028] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties. For example, this system has an interface with relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving consent information from the aforementioned user or organization.

[0029] The present invention will now be described in conjunction with preferred implementation steps. Figure 1 This is a flowchart of a traffic data processing method provided according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0030] Step S101: Based on the historical traffic data of multiple transactions, determine the predicted total traffic cost of multiple transactions within the target time period, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions;

[0031] Step S102: Based on the predicted total traffic cost and system traffic resource data, determine whether to update the traffic parameters of multiple transactions. The system traffic resource data refers to the traffic resource data of the system for processing multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of multiple transactions.

[0032] Step S103: If the predicted total cost of traffic is greater than the system traffic resource data, update the traffic parameters of multiple transactions according to their priorities.

[0033] Through the above steps, based on historical traffic data from multiple transactions, the predicted total traffic cost for multiple transactions within the target time period is determined. This predicted total traffic cost is the sum of the predicted costs for each of the multiple transactions. Based on the predicted total traffic cost and system traffic resource data, it is determined whether to update the traffic parameters for multiple transactions. The system traffic resource data refers to the traffic resources the system uses to process multiple transactions, and the traffic parameters are used to allocate system traffic resources for the traffic of multiple transactions. If the predicted total traffic cost exceeds the system traffic resource data, the traffic parameters for multiple transactions are updated according to their priority. By comparing the predicted traffic cost data for multiple transactions within the target time period with the system traffic resource data, it is determined whether to implement traffic limiting for transactions within the target time period. Then, by adjusting the traffic parameters, traffic limiting is implemented for one or more of the multiple transactions. This achieves the goal of traffic limiting transactions based on their traffic demand at different times, thus improving the utilization rate of system traffic resources. This solves the problem in related technologies where different types of transactions at different times have different traffic demands, but current traffic limiting only considers the type of transaction, resulting in low system throughput and ineffective utilization of system resources.

[0034] The predicted total traffic cost can be the traffic resource data of multiple transactions predicted during the target time period, or the total predicted time cost of the multiple transactions during the target time period. It is used to determine whether to limit the flow of one or more transactions among the multiple transactions in the system during the target time period. The predicted total traffic cost can be predicted using the historical traffic data of the multiple transactions. The prediction model can be trained using the historical traffic data of the multiple transactions. The model can be trained using the transaction time as a parameter and historical transaction volume as sample data until the prediction model converges. The predicted total traffic cost can be determined through the prediction model. The prediction model can be a machine learning model.

[0035] The aforementioned system traffic resource data can be the total time consumed by the maximum number of transactions the system can run over a period of time. For example, if the system can run a maximum of N threads, the total time consumed by the system's running threads in one hour is N × 3600 seconds. By comparing the system traffic resource data with the aforementioned predicted total traffic cost, it can be determined whether the aforementioned multiple transactions need to be rate-limited. If the aforementioned system traffic resource data is greater than or equal to the aforementioned predicted total traffic cost, then the system resources can meet the operation of the aforementioned multiple transactions, and there is no need to rate-limit the aforementioned multiple transactions. If the aforementioned system traffic resource data is less than the aforementioned predicted total traffic cost, and the system resources cannot maintain the original transaction operation state of the aforementioned multiple transactions, then the aforementioned multiple transactions can be rate-limited.

[0036] It should be noted that the rate limiting order for the above-mentioned transactions can be determined based on their priority. The order of priority can be determined from low to high to determine whether to rate limit the transactions. The rate limiting method can be achieved by updating the initial system traffic resource data allocated to the above-mentioned transactions, which is the traffic parameter mentioned above.

[0037] By predicting the traffic cost data of multiple transactions within a target time period and comparing it with system traffic resource data, it is determined whether to limit the flow of transactions during the target time period. Then, by adjusting the traffic parameters, one or more of the multiple transactions are limited, achieving the goal of limiting the flow of transactions according to the traffic demand of different time periods. This achieves the technical effect of improving the utilization rate of system traffic resources, and solves the problem in related technologies where different types of transactions have different traffic demands at different time periods, but current flow limiting only limits the flow based on the type of transaction, resulting in low system throughput and ineffective utilization of system resources.

[0038] Optionally, determining whether to update the traffic parameters of multiple transactions based on the predicted total traffic cost and system traffic resource data further includes: keeping the traffic parameters of multiple transactions unchanged when the predicted total traffic cost is less than or equal to the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority when the predicted total traffic cost is greater than the system traffic resource data, including: determining a first difference between the sum of the traffic costs of multiple transactions and the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority based on the first difference.

[0039] If the predicted total cost of traffic is less than or equal to the system traffic resource data, it indicates that the system traffic resources can meet the requirements of the above multiple transactions with the original traffic parameters, and the traffic parameters of the multiple transactions can remain unchanged; if the predicted total cost of traffic is greater than the system traffic resource data, it indicates that the system traffic resources cannot meet the requirements of the above multiple transactions with the original traffic parameters, and the above traffic parameters can be updated to determine that the above multiple transactions can operate normally, so as to ensure that the system can operate normally.

[0040] When updating the above traffic parameters, we can first determine the first difference between the sum of the traffic costs of multiple transactions and the system traffic resource data. Based on the magnitude of the first difference, we can determine the range of adjustment for the traffic parameters of the above multiple transactions.

[0041] By comparing the predicted total cost of traffic at the target time with the system traffic resource data, it is determined whether to limit the traffic of the above-mentioned multiple transactions at the target time. This achieves the purpose of limiting traffic based on the traffic data demand of transactions at different times, and solves the problem that existing technologies limit transaction traffic only by transaction type, ignoring the impact of time on transaction traffic demand. This achieves the technical effect of improving the utilization rate of system traffic resources.

[0042] Optionally, before updating the traffic parameters of multiple transactions, the method further includes: predicting the total traffic cost based on system traffic resource data and the predicted cost data of each transaction, and determining the initial traffic resource data allocated to each transaction; and using the initial traffic resource data as the traffic parameter of each transaction.

[0043] Before updating the traffic parameters of multiple transactions, the system traffic resource data can be initially allocated. Based on the predicted total traffic cost and the predicted cost data corresponding to each transaction, the initial traffic resource data allocated to each transaction is determined, and the initial traffic resource data is used as the traffic parameter of each transaction.

[0044] For example, suppose the system has a total of n transactions t1...t n By performing the transaction cost function f(t) on each transaction t and sorting the results, we can obtain a descending set of transaction costs Z[f(t1)...f(t)]. n The initial system traffic resources are allocated to each transaction according to the following formula:

[0045]

[0046] Where R represents system traffic resource data, Sum(Z) represents the predicted total transaction cost data for N transactions, and f(t) i ) represents the i-th transaction t i The predicted total cost of traffic, R(t)i The i-th transaction t i The initial traffic resource data, i = 1 to n.

[0047] By predicting the initial traffic resource data for each transaction within a target time period, the goal of predicting the traffic cost of transactions at different times is achieved, thus realizing the technical effect of improving the accuracy of transaction traffic cost prediction.

[0048] Optionally, the traffic parameters include initial traffic resource data. Based on a first difference, the traffic parameters of multiple transactions are updated according to their priorities. This includes: comparing the first predicted cost data of the lowest-priority first transaction with the sum of a preset transaction traffic threshold and a first difference; if the first predicted cost data is greater than the sum of the preset transaction traffic threshold and the first difference, updating the first traffic parameter of the first transaction to the first predicted cost data minus the first difference, and completing the update of the traffic parameters of multiple transactions. The preset transaction traffic threshold is the lowest traffic threshold for multiple transactions to conduct transactions. If the predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to the second difference, where the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; if the second difference is greater than zero, the second predicted cost data of the second transaction among the multiple transactions is compared with the sum of the preset transaction flow threshold and the second difference, where the second transaction is the transaction with the lowest priority among the multiple transactions except for the first transaction; if the second difference is less than or equal to zero, the flow parameters of the multiple transactions are updated.

[0049] The aforementioned transactions can be sorted according to their priority from highest to lowest to form a priority set in ascending or descending order. The preset transaction flow threshold can be the minimum flow threshold for a transaction to be processed.

[0050] The flow parameters of the multiple transactions are updated based on the first difference. This can be achieved by comparing the first predicted cost data of the first transaction with the lowest priority in the priority set with the preset transaction flow threshold. If the first predicted cost data is greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter of the first transaction is updated to the first predicted cost data minus the first difference. This completes the update of the flow parameters of the multiple transactions. In other words, if the first predicted cost data of the first transaction can fully compensate for the part of the total predicted flow cost that exceeds the system flow resource data, including the system resource data of the lowest flow threshold, the flow of the first transaction is directly limited to ensure that the system resources can meet the normal operation of the multiple transactions.

[0051] If the first predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to the second difference, where the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; that is, if the first predicted cost data of the first transaction, while ensuring that the system resource data including the minimum flow threshold is not fully compensated for the part of the total predicted flow cost exceeding the system flow resource data, the first transaction with the lowest priority is allocated only the minimum transaction flow, and the part of the total predicted flow cost exceeding the system flow resource data is updated.

[0052] If the second difference is greater than zero, meaning that the system resources are still insufficient to allow multiple transactions to proceed normally based on the initially allocated traffic resources, then the second predicted cost data of the second transaction among the multiple transactions will be compared with the sum of the preset transaction traffic threshold and the second difference. The second transaction is the transaction with the lowest priority among the multiple transactions except for the first transaction. If the second difference is less than or equal to zero, the traffic parameters of the multiple transactions will be updated.

[0053] By limiting the flow of lower-priority transactions based on the portion of the predicted total flow cost that exceeds the system's flow resource data, the system's flow resources are adjusted according to the predicted transaction costs, ensuring that core transactions are not affected. This achieves the technical effect of improving the balanced allocation rate of the system's transaction flow resources.

[0054] Optionally, before determining the total predicted traffic cost of multiple transactions within a target time period based on the historical traffic data of multiple transactions, the method further includes: acquiring historical transaction data of multiple transactions within a unit of historical time; determining a training sample set based on the transaction time parameters of each transaction within a unit of historical time and the historical transaction volume; training the prediction model based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction within the target time period.

[0055] Before determining the total predicted traffic cost for the target time period based on historical traffic data from multiple transactions, a prediction model can be trained. This prediction model can be a machine learning model. The prediction model can be used to determine the total predicted traffic cost. It can be trained using historical transaction data from multiple transactions within a unit of time, with the transaction time of each transaction as a parameter and the historical transaction volume within a unit of time as a label, until the prediction model converges. The converged prediction model can then be used to predict the transaction volume of each transaction in the target time period.

[0056] For example: Prepare historical transaction data over a period of time. Taking transaction t as an example, use the hour h, month m, and year y from the historical data as parameters, and the transaction volume C of transaction t in that hour as the label to form training data samples. Train this data using a deep learning model M, which has one input layer, two hidden layers, and one output layer. The historical transaction data is then trained using a DNN (Deep Neural Network) algorithm to obtain the prediction model M.

[0057] Optionally, determining the total predicted traffic cost for a target time period based on historical traffic data from multiple transactions includes: obtaining historical traffic data to determine the average time spent on the target transaction, where the target transaction is any one of multiple transactions; determining the target predicted cost data for the target transaction in the target time period based on the average time spent on the target transaction and the prediction model; and determining the total predicted traffic cost for multiple transactions in the target time period based on the target predicted cost data obtained by traversing multiple transactions as target transactions.

[0058] The target transaction can be any one of the multiple transactions mentioned above. Based on the historical traffic data of the time period closest to the target time period in the historical traffic data, the average time spent by the target transaction in the target time period can be determined. Based on the predicted transaction volume of the target transaction in the target time period predicted by the prediction model, and the above average time spent, the target predicted cost data of the target transaction in the target time period can be determined. By traversing multiple transactions as target transactions and obtaining the target predicted cost data, the total predicted traffic cost of multiple transactions in the target time period can be determined, which is to say, the total predicted transaction cost of multiple transactions in the target time period can be determined.

[0059] Optionally, based on the average duration of the target transaction and the prediction model, the target predicted cost data of the target transaction in the target time period is determined. The method further includes: inputting a preset time point of the target transaction in the target time period into the prediction model to obtain the predicted transaction volume of the target transaction in the target time period; and determining the target predicted cost data of the target transaction based on the average duration of the target transaction and the predicted transaction volume.

[0060] Determining the target forecast cost data for the target transaction also includes: inputting the target transaction's preset time points within the target time period into the forecast model to obtain the forecast transaction volume of the target transaction within the target time period; and determining the target forecast cost data for the target transaction based on the average time consumption and forecast transaction volume of the target transaction.

[0061] For example, a trading system performs a flow analysis operation once per hour (i.e., the aforementioned unit of time). The data for the next hour's time point (i.e., the aforementioned preset time point) – year y, month m, and hour h – is fed into the aforementioned deep learning model M (i.e., the aforementioned prediction model) for training, to obtain the predicted trading volume C for the next hour. The average time avg(t) of the transaction t in the current hour is calculated, and then the transaction cost f(t) = avg(t) * C in the next time period can be obtained. The transaction cost in the next time period is also the aforementioned target predicted cost data.

[0062] By using a deep learning model to predict the transaction volume for a target time period, and based on the predicted transaction volume and the average time spent by the target transaction during the target time period, the target predicted cost data for the target transaction is determined. This achieves the goal of using a deep learning model to determine the total cost of the target predicted traffic, and realizes the technical effect of improving the accuracy of transaction cost prediction for the target time period.

[0063] The traffic data processing method provided in this application determines the predicted total traffic cost for a target time period based on historical traffic data of multiple transactions. The predicted total traffic cost is the total cost data of the predicted cost data of multiple transactions. Based on the predicted total traffic cost and system traffic resource data, it determines whether to update the traffic parameters of multiple transactions. The system traffic resource data is the traffic resource data of the processing system for processing multiple transactions, and the traffic parameters are used to allocate resource data for the traffic of multiple transactions. If the traffic parameters of multiple transactions are updated, they are updated according to the priority of the multiple transactions. By comparing the predicted traffic cost data of multiple transactions for the target time period with the system traffic resource data, it is determined whether to limit the flow of transactions during the target time period. Then, by adjusting the traffic parameters, one or more of the multiple transactions are limited in flow. This achieves the purpose of limiting the flow of transactions according to the traffic demand of different time periods, realizing the technical effect of improving the utilization rate of system traffic resources. This solves the problem in related technologies where different types of transactions have different traffic demands at different time periods, but current flow limiting only limits the flow based on the type of transaction, resulting in low system throughput and ineffective utilization of system resources.

[0064] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0065] It should be noted that this application also provides an optional implementation method, which will be described in detail below.

[0066] This implementation provides a precise rate limiting method based on dynamic programming using DNN (Deep Neural Network) algorithms. Since transaction traffic in a production environment has a clear correlation with specific points in time, Figure 2 This is a schematic diagram of the overall architecture of a precise rate limiting method based on DNN dynamic programming according to an embodiment of this application. The overall architecture to be adopted in this embodiment is as follows: Figure 2 As shown:

[0067] This precise traffic limiting solution consists of an offline training module and an online recognition module;

[0068] Offline training module: Prepare historical transaction data over a period of time. Taking transaction t as an example, use the hour h, month m, and year y from the historical data as parameters (i.e., the time parameters mentioned above), and the transaction volume C of transaction t in that hour as the label to form training data samples. Train the data using a deep learning model M (i.e., the prediction model mentioned above), which has one input layer, two hidden layers, and one output layer. Train the historical transaction data using a DNN to obtain the prediction model M.

[0069] The online analysis module first assesses the total system resources (i.e., the system traffic resource data mentioned above) R of a system within one hour, using the assessment method R = tps * 3600, where tps is the maximum tps that can support the system's stable operation over a long period. Then, it assesses the importance of transactions within the system, obtaining the ascending priority set E[t1...t...]. n A trading system performs a flow analysis operation once per hour, feeding the time data y, m, h of the next hour into the deep learning model M for training to obtain the predicted trading volume C for the next hour. The system then calculates the average time avg(t) of the current hour for the transaction t, and finally obtains the transaction cost f(t) = avg(t) * C in the next time period. Assume the system has a total of n transactions t1...t n By performing the transaction cost function f(t) on each transaction t and sorting the results, we can obtain a descending set of transaction costs Z[f(t1)...f(t)]. n The initial system traffic resources are allocated to each transaction according to the following formula:

[0070]

[0071] The dynamic precise rate limiting algorithm is shown in Table 1:

[0072] Table 1

[0073]

[0074] The key to this implementation is that by introducing a machine learning model into the rate limiting scenario, the maximum rate limiting resources that each transaction can occupy can be dynamically allocated over time, in order to maximize the utilization of system resources.

[0075] This implementation introduces a deep learning model to predict transaction volume over a future period and proposes a dynamic flow-limiting algorithm. This algorithm can balance and allocate system resources based on the predicted transaction volume, so as to automatically degrade non-core transactions and protect core transactions from being affected in the event of sudden abnormal traffic.

[0076] This application also provides a traffic data processing apparatus. It should be noted that the traffic data processing apparatus of this application can be used to execute the traffic data processing method provided in this application. The traffic data processing apparatus provided in this application will be described below.

[0077] Figure 3 This is a schematic diagram of a traffic data processing device according to an embodiment of this application, such as... Figure 3 As shown, the device includes: a first determining module 30, a second determining module 32, and an updating module 34. The device will be described in detail below.

[0078] The traffic data processing device includes a first determining module 30, which determines the predicted total traffic cost of multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; a second determining module 32, connected to the first determining module 30, which determines whether to update the traffic parameters of multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the resource data of the processing system for processing transaction traffic, and the traffic parameters are used to allocate resource data for the traffic of multiple transactions; and an updating module 34, connected to the second determining module 32, which updates the traffic parameters of multiple transactions according to the priority of the multiple transactions when the predicted total traffic cost is greater than the system traffic resource data.

[0079] The traffic data processing apparatus provided in this application includes a first determining module 30 that determines the predicted total traffic cost of multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; a second determining module 32 that determines whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the resource data of the processing system for processing transaction traffic, and the traffic parameters are used to allocate resource data for the traffic of the multiple transactions; and an updating module 34 that updates the traffic parameters of the multiple transactions according to the priority of the multiple transactions if the predicted total traffic cost is greater than the system traffic resource data. By predicting the traffic cost data of multiple transactions within a target time period and comparing it with system traffic resource data, it is determined whether to limit the flow of transactions during the target time period. Then, by adjusting the traffic parameters, one or more of the multiple transactions are limited, achieving the goal of limiting the flow of transactions according to the traffic demand of different time periods. This achieves the technical effect of improving the utilization rate of system traffic resources, and solves the problem in related technologies where different types of transactions have different traffic demands at different time periods, but current flow limiting only limits the flow based on the type of transaction, resulting in low system throughput and ineffective utilization of system resources.

[0080] Optionally, it also includes: a judgment unit, which keeps the traffic parameters of multiple transactions unchanged when the predicted total traffic cost is less than or equal to the system traffic resource data; and an update unit, which determines the first difference between the sum of the traffic costs of multiple transactions and the system traffic resource data; and updates the traffic parameters of multiple transactions according to the priority of the multiple transactions based on the first difference.

[0081] Optionally, before updating the traffic parameters of multiple transactions, the method further includes: predicting the total traffic cost based on system traffic resource data and the predicted cost data for each transaction, and determining the initial traffic resource data allocated to each transaction; and using the initial traffic resource data as the traffic parameter for each transaction.

[0082] Optionally, the second judgment unit includes: comparing the first predicted cost data of the lowest priority first transaction among multiple transactions with the sum of a preset transaction flow threshold and a first difference; if the first predicted cost data is greater than the sum of the preset transaction flow threshold and the first difference, updating the first flow parameter of the first transaction to the first predicted cost data minus the first difference, and completing the update of the flow parameters of multiple transactions, wherein the preset transaction flow threshold is the lowest flow threshold for multiple transactions to conduct transactions; if the first predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, updating the first flow parameter to the preset transaction flow threshold; updating the first difference to a second difference, wherein the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; if the second difference is greater than zero, continuing to compare the second predicted cost data of the second transaction among multiple transactions with the sum of the preset transaction flow threshold and the second difference, wherein the second transaction is the lowest priority transaction among multiple transactions excluding the first transaction; if the second difference is less than or equal to zero, completing the update of the flow parameters of multiple transactions.

[0083] Optionally, it also includes: an acquisition module, used to acquire historical transaction data of multiple transactions within a unit historical time, and determine a training sample set based on the transaction time parameter of each transaction within a unit historical time and the historical transaction volume; and a training module, used to train the prediction model based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction within a target time period.

[0084] Optionally, the first determining module includes: an acquisition unit, which acquires historical traffic data to determine the average time consumption of the target transaction, wherein the target transaction is any one of multiple transactions; based on the average time consumption of the target transaction and the prediction model, the first determining unit determines the target predicted cost data of the target transaction in the target time period; and a second determining unit, which determines the total predicted traffic cost of the multiple transactions in the target time period based on the target predicted cost data obtained by traversing multiple transactions as target transactions respectively.

[0085] Optionally, the first determining unit includes: inputting a preset time point of the target transaction within a target time period into the prediction model to obtain the predicted transaction volume of the target transaction in the target time period; and determining the target prediction cost data of the target transaction based on the average time consumption and the predicted transaction volume of the target transaction.

[0086] The traffic data processing device includes a processor and a memory. The first determining module 30, the second determining module 32, the updating module 36, etc., are all stored in the memory as program units. The processor executes the program units stored in the memory to realize the corresponding functions.

[0087] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured. By adjusting kernel parameters, the varying transaction flow requirements of different types of transactions at different times in related technologies can be addressed. However, current rate limiting only applies to the type of transaction, resulting in low system throughput and inefficient utilization of system resources.

[0088] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0089] This invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the traffic data processing method.

[0090] This invention provides a processor for running a program, wherein the program executes the traffic data processing method during runtime.

[0091] Figure 4 This is a schematic diagram of an electronic device provided according to an embodiment of this application, such as... Figure 4 As shown, this application embodiment provides an electronic device 40, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: determining the predicted total traffic cost of multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; determining whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the traffic resource data of the system for processing multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; and updating the traffic parameters of the multiple transactions according to the priority of the multiple transactions if the predicted total traffic cost is greater than the system traffic resource data.

[0092] Optionally, determining whether to update the traffic parameters of multiple transactions based on the predicted total traffic cost and system traffic resource data further includes: keeping the traffic parameters of multiple transactions unchanged when the predicted total traffic cost is less than or equal to the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority when the predicted total traffic cost is greater than the system traffic resource data, including: determining a first difference between the sum of the traffic costs of multiple transactions and the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority based on the first difference.

[0093] Optionally, before updating the traffic parameters of multiple transactions, the method further includes: predicting the total traffic cost based on system traffic resource data and the predicted cost data of each transaction, and determining the initial traffic resource data allocated to each transaction; and using the initial traffic resource data as the traffic parameter of each transaction.

[0094] Optionally, the traffic parameters include initial traffic resource data. Based on a first difference, the traffic parameters of multiple transactions are updated according to their priorities. This includes: comparing the first predicted cost data of the lowest-priority first transaction with the sum of a preset transaction traffic threshold and a first difference; if the first predicted cost data is greater than the sum of the preset transaction traffic threshold and the first difference, updating the first traffic parameter of the first transaction to the first predicted cost data minus the first difference, and completing the update of the traffic parameters of multiple transactions. The preset transaction traffic threshold is the lowest traffic threshold for multiple transactions to conduct transactions. If the predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to the second difference, where the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; if the second difference is greater than zero, the second predicted cost data of the second transaction among the multiple transactions is compared with the sum of the preset transaction flow threshold and the second difference, where the second transaction is the transaction with the lowest priority among the multiple transactions except for the first transaction; if the second difference is less than or equal to zero, the flow parameters of the multiple transactions are updated.

[0095] Optionally, before determining the total predicted traffic cost of multiple transactions within a target time period based on the historical traffic data of multiple transactions, the method further includes: acquiring historical transaction data of multiple transactions within a unit of historical time; determining a training sample set based on the transaction time parameters of each transaction within a unit of historical time and the historical transaction volume; training the prediction model based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction within the target time period.

[0096] Optionally, determining the total predicted traffic cost for a target time period based on historical traffic data from multiple transactions includes: obtaining historical traffic data to determine the average time spent on the target transaction, where the target transaction is any one of multiple transactions; determining the target predicted cost data for the target transaction in the target time period based on the average time spent on the target transaction and the prediction model; and determining the total predicted traffic cost for multiple transactions in the target time period based on the target predicted cost data obtained by traversing multiple transactions as target transactions.

[0097] Optionally, based on the average duration of the target transaction and the prediction model, the target predicted cost data of the target transaction in the target time period is determined. The method further includes: inputting a preset time point of the target transaction in the target time period into the prediction model to obtain the predicted transaction volume of the target transaction in the target time period; and determining the target predicted cost data of the target transaction based on the average duration of the target transaction and the predicted transaction volume.

[0098] The devices mentioned in this article can be servers, PCs, tablets, mobile phones, etc.

[0099] This application also provides a computer program product, which, when executed on a traffic data processing device, is suitable for executing an initialization program having the following method steps: determining the predicted total traffic cost of multiple transactions within a target time period based on historical traffic data of multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; determining whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data, wherein the system traffic resource data is the traffic resource data of the system for processing multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; and updating the traffic parameters of the multiple transactions according to the priority of the multiple transactions if the predicted total traffic cost is greater than the system traffic resource data.

[0100] Optionally, determining whether to update the traffic parameters of multiple transactions based on the predicted total traffic cost and system traffic resource data further includes: keeping the traffic parameters of multiple transactions unchanged when the predicted total traffic cost is less than or equal to the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority when the predicted total traffic cost is greater than the system traffic resource data, including: determining a first difference between the sum of the traffic costs of multiple transactions and the system traffic resource data; and updating the traffic parameters of multiple transactions according to their priority based on the first difference.

[0101] Optionally, before updating the traffic parameters of multiple transactions, the method further includes: predicting the total traffic cost based on system traffic resource data and the predicted cost data of each transaction, and determining the initial traffic resource data allocated to each transaction; and using the initial traffic resource data as the traffic parameter of each transaction.

[0102] Optionally, the traffic parameters include initial traffic resource data. Based on a first difference, the traffic parameters of multiple transactions are updated according to their priorities. This includes: comparing the first predicted cost data of the lowest-priority first transaction with the sum of a preset transaction traffic threshold and a first difference; if the first predicted cost data is greater than the sum of the preset transaction traffic threshold and the first difference, updating the first traffic parameter of the first transaction to the first predicted cost data minus the first difference, and completing the update of the traffic parameters of multiple transactions. The preset transaction traffic threshold is the lowest traffic threshold for multiple transactions to conduct transactions. If the predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to the second difference, where the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold; if the second difference is greater than zero, the second predicted cost data of the second transaction among the multiple transactions is compared with the sum of the preset transaction flow threshold and the second difference, where the second transaction is the transaction with the lowest priority among the multiple transactions except for the first transaction; if the second difference is less than or equal to zero, the flow parameters of the multiple transactions are updated.

[0103] Optionally, before determining the total predicted traffic cost of multiple transactions within a target time period based on the historical traffic data of multiple transactions, the method further includes: acquiring historical transaction data of multiple transactions within a unit of historical time; determining a training sample set based on the transaction time parameters of each transaction within a unit of historical time and the historical transaction volume; training the prediction model based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction within the target time period.

[0104] Optionally, determining the total predicted traffic cost for a target time period based on historical traffic data from multiple transactions includes: obtaining historical traffic data to determine the average time spent on the target transaction, where the target transaction is any one of multiple transactions; determining the target predicted cost data for the target transaction in the target time period based on the average time spent on the target transaction and the prediction model; and determining the total predicted traffic cost for multiple transactions in the target time period based on the target predicted cost data obtained by traversing multiple transactions as target transactions.

[0105] Optionally, based on the average duration of the target transaction and the prediction model, the target predicted cost data of the target transaction in the target time period is determined. The method further includes: inputting a preset time point of the target transaction in the target time period into the prediction model to obtain the predicted transaction volume of the target transaction in the target time period; and determining the target predicted cost data of the target transaction based on the average duration of the target transaction and the predicted transaction volume.

[0106] 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.

[0107] 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 traffic data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable traffic data processing device, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0108] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable traffic data processing device to operate 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.

[0109] These computer program instructions can also be loaded onto a computer or other programmable traffic data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device 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.

[0110] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0111] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0112] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0113] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0114] 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.

[0115] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for processing traffic data, characterized in that, include: Based on historical traffic data of multiple transactions, the predicted total traffic cost of the multiple transactions within a target time period is determined, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; Based on the predicted total traffic cost and system traffic resource data, it is determined whether to update the traffic parameters of the multiple transactions, wherein the system traffic resource data is the traffic resource data of the system processing the multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions; If the predicted total cost of traffic is greater than the system traffic resource data, the traffic parameters of the multiple transactions are updated according to their priorities, including: determining a first difference between the sum of the traffic costs of the multiple transactions and the system traffic resource data; and updating the traffic parameters of the multiple transactions according to their priorities based on the first difference.

2. The method according to claim 1, characterized in that, After determining whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and the system traffic resource data, the method further includes: If the total predicted traffic cost is less than or equal to the system traffic resource data, the traffic parameters of the multiple transactions remain unchanged.

3. The method according to claim 2, characterized in that, Before updating the flow parameters of the plurality of transactions, the method further includes: Based on the system traffic resource data, the predicted total traffic cost, and the predicted cost data for each transaction, the initial traffic resource data allocated to each transaction is determined respectively. The initial traffic resource data is used as the traffic parameter for each transaction.

4. The method according to claim 3, characterized in that, The traffic parameters include the initial traffic resource data. Based on the first difference, the traffic parameters of the multiple transactions are updated according to their priority, including: The first predicted cost data of the first transaction with the lowest priority among the multiple transactions is compared with the sum of a preset transaction flow threshold and the first difference. If the first predicted cost data is greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter of the first transaction is updated to the first predicted cost data minus the first difference, and the flow parameters of the plurality of transactions are updated. The preset transaction flow threshold is the minimum flow threshold for the plurality of transactions to conduct transactions. If the first predicted cost data is not greater than the sum of the preset transaction flow threshold and the first difference, the first flow parameter is updated to the preset transaction flow threshold; the first difference is updated to a second difference, wherein the second difference is the first difference minus the difference between the first predicted cost data and the preset transaction flow threshold. If the second difference is greater than zero, the second predicted cost data of the second transaction among the plurality of transactions is further compared with the sum of the preset transaction flow threshold and the second difference, wherein the second transaction is the transaction with the lowest priority among the plurality of transactions except the first transaction; If the second difference is less than or equal to zero, the flow parameters of the multiple transactions are updated.

5. The method according to claim 1, characterized in that, Before determining the predicted total cost of traffic for the multiple transactions within a target time period based on historical traffic data from multiple transactions, the method further includes: Obtain historical transaction data for the multiple transactions within a unit of historical time. The training sample set is determined based on the transaction time parameter of each transaction within the unit of historical time and the historical transaction volume. The prediction model is trained based on the training sample set until the prediction model converges, wherein the prediction model is used to predict the transaction volume of a single transaction in the target time period.

6. The method according to claim 5, characterized in that, Based on historical traffic data from multiple transactions, the predicted total traffic cost for the target time period is determined, including: The average time taken to determine the target transaction is determined by acquiring the historical traffic data, wherein the target transaction is any one of the plurality of transactions; Based on the average time taken for the target transaction and the prediction model, the target predicted cost data for the target transaction in the target time period is determined; Based on the target predicted cost data obtained by traversing the multiple transactions as the target transactions, the total predicted traffic cost of the multiple transactions in the target time period is determined.

7. The method according to claim 6, characterized in that, Based on the average time elapsed for the target transaction and the prediction model, the method further includes determining the target predicted cost data for the target transaction during the target time period. The target transaction is input into the prediction model at a preset time point within the target time period to obtain the predicted transaction volume of the target transaction within the target time period; The target predicted cost data for the target transaction is determined based on the average time taken for the target transaction and the predicted transaction volume.

8. A flow data processing device, characterized in that, include: The first determining module determines the predicted total traffic cost of the multiple transactions within a target time period based on the historical traffic data of the multiple transactions, wherein the predicted total traffic cost is the sum of the predicted cost data of each of the multiple transactions; The second determining module determines whether to update the traffic parameters of the multiple transactions based on the predicted total traffic cost and system traffic resource data. The system traffic resource data refers to the traffic resource data of the system for processing the multiple transactions, and the traffic parameters are used to allocate system traffic resource data for the traffic of the multiple transactions. An update module is used to update the traffic parameters of the multiple transactions according to their priorities when the predicted total traffic cost is greater than the system traffic resource data. The update unit determines a first difference between the sum of the traffic costs of the multiple transactions and the system traffic resource data; based on the first difference, it updates the traffic parameters of the multiple transactions according to their priority.

9. A computer-readable storage medium, characterized in that, The storage medium is used to store a program, wherein the program executes the traffic data processing method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the traffic data processing method according to any one of claims 1 to 7.

Images