Resource transfer channel control method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a resource transfer channel control method and device, a storage medium and electronic equipment, and the method comprises the steps: carrying out the analysis of resource transfer channel data in a plurality of historical time ranges through a pre-trained dynamic fusing decision model, and obtaining a fusing decision result and a prediction recovery strategy; the prediction recovery strategy is used for specifying the recovery opportunity and the flow recovery rate of the current resource transfer channel; the recovery opportunity refers to a time sequence for recovering the current resource transfer channel after a preset duration after fusing the current resource transfer channel; the flow recovery rate refers to the flow increase rate in the process of recovering the current resource transfer channel; and under the condition that the fusing decision result represents fusing of the current resource transfer channel, fusing the current resource transfer channel, and recovering the current resource transfer channel at the moment indicated by the recovery opportunity according to the flow recovery rate. According to the invention, the problems of low efficiency and high misjudgment rate caused by manual intervention of fusing recovery are solved.

Description

Technical Field

[0001] This application relates to the field of computers, and more specifically, to a resource transfer channel control method, apparatus, storage medium, and electronic device. Background Technology

[0002] The resource transfer channel is a key component connecting resource platforms and resource management institutions. It is responsible for processing user resource transfer requests and performing operations such as transaction authorization and settlement. The resource transfer channel may malfunction due to various reasons, such as system upgrades or maintenance, network jitter or latency, internal failures of the resource system, and high concurrency pressure, requiring the resource transfer channel to be tripped.

[0003] The existing circuit breaker scheme requires manual intervention to restore the circuit breaker after the circuit breaker resources are transferred to the channel, which is inefficient and has the risk of misjudgment. Summary of the Invention

[0004] This application provides a resource transfer channel control method, apparatus, storage medium, and electronic device to at least solve the technical problems of low efficiency and high misjudgment rate caused by manual intervention in circuit breaker recovery in related technologies.

[0005] According to one aspect of the embodiments of this application, a resource transfer channel control method is provided, comprising: acquiring resource transfer channel data of a current resource transfer channel within multiple consecutive historical time ranges; analyzing the resource transfer channel data within the multiple historical time ranges through a pre-trained dynamic circuit breaker decision model to obtain a circuit breaker decision result and a predicted recovery strategy; the predicted recovery strategy is used to specify the recovery timing and traffic recovery rate of the current resource transfer channel; the recovery timing refers to the timing of restoring the current resource transfer channel after a preset time period following the circuit breaker decision; the traffic recovery rate refers to the rate of increase of traffic during the restoration of the current resource transfer channel; when the circuit breaker decision result indicates that the current resource transfer channel is to be circuit breaker-broken, the current resource transfer channel is to be circuit-breaker-broken, the traffic of the current resource transfer channel is reduced to a first traffic level, and the current resource transfer channel is restored at the time indicated by the recovery timing according to the traffic recovery rate.

[0006] According to another aspect of the embodiments of this application, a resource transfer channel control device is also provided, comprising: a data acquisition module, configured to acquire resource transfer channel data of the current resource transfer channel within multiple consecutive historical time ranges; a dynamic decision-making module, configured to analyze the resource transfer channel data within the multiple historical time ranges through a pre-trained dynamic circuit breaker decision model to obtain a circuit breaker decision result and a predicted recovery strategy; the predicted recovery strategy is used to specify the recovery timing and traffic recovery rate of the current resource transfer channel; the recovery timing refers to the timing of restoring the current resource transfer channel after a preset time period following the circuit breaker decision; the traffic recovery rate refers to the rate of increase of traffic during the restoration of the current resource transfer channel; and a channel control module, configured to, when the circuit breaker decision result indicates that the current resource transfer channel has been circuit breakered, circuit breaker the current resource transfer channel, reduce the traffic of the current resource transfer channel to a first traffic level, and restore the current resource transfer channel at the time indicated by the recovery timing according to the traffic recovery rate.

[0007] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to perform the steps in any of the above method embodiments when it is run.

[0008] According to another aspect of the embodiments of this application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the steps in any of the method embodiments described above.

[0009] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to perform the steps of any of the above method embodiments through the computer program.

[0010] This application utilizes a pre-trained dynamic circuit breaker decision model to analyze resource transfer channel data across multiple historical time ranges, obtaining circuit breaker decision results and predictive recovery strategies. The predictive recovery strategy not only intelligently predicts the recovery timing of resource transfer channels but also guides the gradual recovery rate of traffic, avoiding the blindness and lag that may exist in manual recovery, significantly improving recovery efficiency and channel stability. Combining the dynamic circuit breaker decision results and predictive recovery strategies, automatic circuit breaker and intelligent recovery of resource transfer channels are achieved. Once it is determined that the current resource transfer channel needs to be circuit-broken, the circuit breaker is automatically triggered, reducing traffic to a safe first level. Subsequently, based on the predictive recovery strategy, channel traffic can be automatically restored at an appropriate time and with a suitable recovery rate. The entire process requires no manual intervention, reducing labor costs and avoiding errors that may occur with manual judgment. This solves the technical problems of low efficiency and high misjudgment rate caused by manual intervention in circuit breaker recovery in related technologies, improving the efficiency and accuracy of circuit breaker and recovery, and reducing the risk of misjudgment. Attached Figure Description

[0011] Figure 1 This is a schematic diagram illustrating an application scenario of a resource transfer channel control method according to an embodiment of this application;

[0012] Figure 2 This is a flowchart illustrating an optional resource transfer channel control method according to an embodiment of this application;

[0013] Figure 3 This is a schematic diagram of an optional training method for a dynamic circuit breaker decision model according to an embodiment of this application;

[0014] Figure 4 This is a schematic diagram of another optional dynamic circuit breaker decision model according to an embodiment of this application;

[0015] Figure 5 This is a structural block diagram of an optional resource transfer channel control device according to an embodiment of this application;

[0016] Figure 6 This is a computer system architecture block diagram of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0017] 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.

[0018] 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 so that the embodiments of this application described herein can be implemented in orders other than those illustrated or 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.

[0019] According to one aspect of the embodiments of this application, a resource transfer channel control method is provided. Optionally, in this embodiment, the above-described resource transfer channel control method may be applied to, but is not limited to, [examples of applications]. Figure 1 The hardware environment shown includes terminal device 102 and server 104. Server 104 can be connected to terminal device 102 via a network and can be used to provide services (e.g., application services, etc.) to terminal device 102 or clients installed on terminal device 102. A database can be set up on server 104 or independently of server 104 to provide data storage services for server 104.

[0020] The aforementioned network may include, but is not limited to, at least one of the following: wired network and wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network (WAN), metropolitan area network (MAN), and local area network (LAN). The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity) and Bluetooth. Terminal device 102 may be, but is not limited to, PC (Personal Computer), mobile phone, tablet computer, etc. Server 104 may be, but is not limited to, cloud server, server cluster, or other server types.

[0021] The resource transfer channel control method of this application embodiment can be executed by server 104, terminal device 102, or jointly by server 104 and terminal device 102. Alternatively, the resource transfer channel control method of this application embodiment can be executed by a client installed on terminal device 102.

[0022] Taking the resource transfer channel control method in this embodiment as an example, which is executed by server 104, Figure 2 This is a flowchart illustrating an optional resource transfer channel control method according to an embodiment of this application, as shown below. Figure 2As shown, the process of this method may include the following steps:

[0023] Step S202: Obtain resource transfer channel data for the current resource transfer channel within multiple consecutive historical time ranges.

[0024] The resource transfer channel control method in this embodiment can be applied to high-concurrency, highly volatile online service environments, and is particularly suitable for scenarios with strict requirements on data processing speed, accuracy, and continuity. Examples include e-commerce resource transfer, financial transaction resource transfer, large-scale online game resource transfer, and cloud computing resource allocation.

[0025] In a resource transfer system, a resource transfer channel refers to the communication path or processing unit connecting the resource transfer initiator and receiver, responsible for data transmission and verification. The current resource transfer channel refers to the resource transfer channel currently being evaluated among multiple resource transfer channels. Resource isolation is implemented for multiple resource transfer channels to ensure that circuit breaking operations do not affect other services. Specifically, the circuit breaking operation of each resource transfer channel is restricted to an independent container to avoid affecting other resource transfer channels. For example, the circuit breaking operation of resource transfer channel A is restricted to an independent container to avoid affecting resource transfer channel B.

[0026] Resource transfer channel data refers to various indicators that reflect the health status and operational efficiency of the resource transfer channel, such as, but not limited to, resource transfer success rate, real-time transfer volume (TPS) in processing, average response time, and error code distribution.

[0027] Historical time ranges refer to specified time periods used to analyze the historical performance of resource transfer channels, helping to identify long-term patterns and anomalous behaviors. Multiple consecutive historical time ranges refer to several consecutive and non-overlapping time periods used to gain a more comprehensive understanding of the operational status of resource transfer channels.

[0028] Step S204: Analyze resource transfer channel data across multiple historical time ranges using a pre-trained dynamic circuit breaker decision model to obtain circuit breaker decision results and predicted recovery strategies. The predicted recovery strategy specifies the recovery timing and traffic recovery rate of the current resource transfer channel. The recovery timing refers to the timing of restoring the current resource transfer channel after a preset time period following the circuit breaker termination. The traffic recovery rate refers to the rate at which traffic increases during the restoration of the current resource transfer channel.

[0029] The dynamic circuit breaker decision model refers to a model pre-trained using deep learning techniques (such as LSTM (Long Short-Term Memory) and reinforcement learning) to analyze the health status of resource transfer channels in real time and dynamically generate circuit breaker thresholds. Faced with the dilemma of fixed circuit breaker thresholds failing due to business fluctuations, the dynamic circuit breaker decision model can combine historical data and real-time indicators to predict channel health trends and intelligently decide when to trigger a circuit breaker, avoiding inappropriate circuit breaking that causes service interruptions or excessive tolerance of anomalies that leads to excessive system pressure.

[0030] After analyzing resource transfer channel data, the dynamic circuit breaker decision model arrives at a conclusion on whether the circuit breaker mechanism should be activated, along with specific parameter settings for the circuit breaker (such as success rate thresholds). This circuit breaker decision directly informs the intelligent circuit breaker operation. The dynamic circuit breaker decision model dynamically adjusts the circuit breaker threshold based on the channel's real-time performance, ensuring action is taken at the most appropriate time. This effectively avoids unnecessary interruptions to channel services and prevents secondary circuit breakers that may result from premature recovery.

[0031] Predictive recovery strategies, generated by dynamic circuit breaker decision models, guide resource transfer channels on how to restore service, including the timing of recovery and the planning of traffic ramp-up rates. These strategies not only predict when it's appropriate to begin channel recovery but also define the rate of traffic recovery. Through gradual recovery mechanisms, such as initial liveness testing and subsequent linear traffic recovery at a rate of 5% per minute, the strategies ensure that the channel can quickly return to normal during recovery while avoiding new stability issues caused by sudden traffic surges.

[0032] Predictive recovery strategies specify the timing and rate of traffic recovery for the current resource transfer channel. The recovery timing refers to the optimal time to restore service after a circuit breaker event, as determined by the predictive recovery strategy. After a certain period (e.g., 5 minutes) following a circuit breaker event, the system waits for the appropriate time (based on a liveness test) before determining whether to begin restoring channel traffic. This delayed recovery strategy considers the channel's actual recovery capacity and potential risks, avoiding instability that might result from premature recovery. The traffic recovery rate is a control parameter that controls the speed at which traffic gradually recovers to normal levels during the circuit breaker recovery process, affecting the smoothness and efficiency of channel recovery.

[0033] Optionally, the server collects resource transfer channel data for the current resource transfer channel across multiple consecutive historical time ranges, and inputs this data into the dynamic circuit breaker decision model. The dynamic circuit breaker decision model analyzes the resource transfer channel data across multiple consecutive historical time ranges, dynamically calculates the circuit breaker threshold, assesses whether the current resource transfer channel's circuit breaker index exceeds the corresponding circuit breaker threshold, and outputs the circuit breaker decision result (whether to suspend the circuit breaker) and the predicted recovery strategy based on the assessment results.

[0034] Step S206: If the circuit breaker decision result indicates that the current resource transfer channel is to be broken, the current resource transfer channel is broken, the flow of the current resource transfer channel is reduced to the first flow, and the current resource transfer channel is restored at the time indicated by the restoration timing, according to the flow restoration rate.

[0035] The circuit breaker decision result indicates that the current resource transfer channel is being suspended, meaning the dynamic circuit breaker decision model outputs a "yes" decision. In this case, the current resource transfer channel is suspended, and its flow is reduced to a minimum safe level. The minimum safe level is the lowest flow rate the channel flow is reduced to when the circuit breaker is activated, to prevent further escalation of the anomaly. For example, during the execution of the circuit breaker operation, the channel flow is reduced to 1% to ensure that rapid action can be taken to minimize the impact in the event of a channel anomaly.

[0036] Optionally, if the circuit breaker decision indicates that the current resource transfer channel is being broken, the server breaks the current resource transfer channel, reduces the traffic of the current resource transfer channel to the first level, and records the circuit breaker time and reason. When the recovery time arrives, the server begins to gradually increase the channel traffic according to the predicted traffic recovery rate to ensure channel stability until the traffic is fully restored.

[0037] The resource transfer channel control method described in this application, evaluated from three indicators—fault false alarm rate, anomaly recovery time, and number of manual interventions—is an improvement over existing circuit breaker schemes. Table 1 shows the measured values ​​of these three indicators when using the resource transfer channel control method of this application and existing circuit breaker schemes in the same scenario:

[0038] Table 1

[0039] index Existing circuit breaker scheme Examples of this application Improvement effect Circuit breaker misjudgment rate 25％ 8％ Reduced by 67% Abnormal recovery time 15 minutes 3 minutes shorten by 80% Number of manual interventions 10 times / day 1 time / day Reduce by 90%

[0040] Therefore, compared with existing circuit breaker schemes, the resource transfer channel control method of this application reduces the circuit breaker false alarm rate by 67%, improves the abnormal recovery time by 80% (from an average of 15 minutes to 3 minutes), and reduces the need for manual intervention by 90%. It is evident that the resource transfer channel control method of this application has significant progress compared with existing circuit breaker schemes.

[0041] The embodiments provided in this application analyze resource transfer channel data across multiple historical time ranges using a pre-trained dynamic circuit breaker decision model to obtain circuit breaker decision results and predictive recovery strategies. The predictive recovery strategy not only intelligently predicts the recovery timing of resource transfer channels but also guides the gradual recovery rate of traffic, avoiding the blindness and lag that may exist in manual recovery, greatly improving recovery efficiency and channel stability. Combining the dynamic circuit breaker decision results and predictive recovery strategies, automatic circuit breaker and intelligent recovery of resource transfer channels are achieved. Once it is determined that the current resource transfer channel needs to be circuit-breakered, the circuit breaker is automatically triggered, reducing traffic to a safe first flow rate. Then, based on the predictive recovery strategy, channel traffic is automatically restored at an appropriate time and with a suitable recovery rate. The entire process requires no manual intervention, reducing labor costs and avoiding errors that may occur with manual judgment. This solves the technical problems of low efficiency and high misjudgment rate caused by manual intervention in circuit breaker recovery in related technologies, improving the efficiency and accuracy of circuit breaker and recovery, and reducing the risk of misjudgment.

[0042] In one exemplary embodiment, a pre-trained dynamic circuit breaker decision model is used to analyze resource transfer channel data across multiple historical timeframes to obtain circuit breaker decision results, including:

[0043] First, by using a pre-trained dynamic circuit breaker decision model, we analyze the resource transfer channel data across multiple historical time ranges to obtain the health index of the current resource transfer channel in each historical time range, as well as the dynamic circuit breaker threshold of the current resource transfer channel in each historical time range.

[0044] Existing circuit breaker schemes typically rely on fixed thresholds for judgment. For example, when the transfer success rate drops below 90%, the system automatically triggers a circuit breaker, limiting traffic to the resource transfer channel to prevent further system deterioration or crashes. However, fixed thresholds cannot adapt to fluctuations in traffic volume. During off-peak periods, anomalies in the resource transfer channel may be overlooked, while during peak periods, even minor anomalies may trigger a circuit breaker, causing unnecessary service interruptions. To address this issue, this embodiment utilizes a pre-trained dynamic circuit breaker decision model, combined with an LSTM time-series prediction model and reinforcement learning algorithms, to dynamically generate circuit breaker thresholds. This means that circuit breaker decisions are no longer static but can automatically adjust based on current traffic fluctuations and channel status, significantly improving the flexibility and accuracy of the circuit breaker mechanism and resolving the problems of fixed thresholds and poor adaptability in traditional circuit breaker schemes.

[0045] The health index is a quantitative indicator of the operational status of a resource transfer channel within a specific time frame. It combines data from multiple dimensions, including success rate, TPS, error code frequency, and transaction volume, and is calculated using a weighted formula. A healthy channel should maintain a high health index, while a sustained drop below a threshold may trigger a circuit breaker mechanism.

[0046] Dynamic circuit breaker thresholds are triggering standards that are dynamically generated based on the actual operating environment and historical data of the resource transfer channel, using deep learning models (such as LSTM) and reinforcement learning algorithms. In traditional fixed-threshold circuit breaker schemes, threshold settings are often applied uniformly, while dynamic circuit breaker thresholds take into account business fluctuations (such as peak and off-peak periods), channel characteristics, and historical anomaly patterns, making circuit breaker decisions more accurate and efficient. For example, during peak periods, the threshold may be lowered to avoid premature circuit breaking, while during off-peak periods, the threshold may be raised to reduce the tolerance for service interruptions.

[0047] Optionally, the server performs data cleaning on resource transfer channel data across multiple consecutive historical time ranges, removes missing values, and standardizes the numerical range to obtain processed resource transfer channel data across multiple historical time ranges. This processed data is then input into a dynamic circuit breaker decision model. The model analyzes the resource transfer channel data for each historical time range to obtain the health index corresponding to each historical time range and the dynamic circuit breaker threshold for the current resource transfer channel in each historical time range.

[0048] Second, when the health index is less than the corresponding dynamic circuit breaker threshold in each historical time range, the dynamic circuit breaker decision model outputs the circuit breaker decision result that represents the circuit breaker decision to suspend the current resource transfer channel.

[0049] Optionally, if the health index is less than the corresponding dynamic circuit breaker threshold in each historical time range, it is determined that the current resource transfer channel needs to be circuit breaker-broken, and the server outputs the circuit breaker decision result representing the circuit breaker decision to break the current resource transfer channel through the dynamic circuit breaker decision model.

[0050] In some embodiments, Figure 3 This is a schematic diagram of an optional training method for a dynamic circuit breaker decision model according to an embodiment of this application; as shown. Figure 3As shown, the dynamic circuit breaker decision model includes a data acquisition module, a training module, and a decision module. The model training process includes the following steps: The data acquisition module collects historical transaction data (including real-time acquisition success rate, processing transaction volume, error code types, and frequencies in normal and abnormal scenarios); the error code semantic analysis module (NLP) identifies key anomaly types and integrates historical manual circuit breaker records, labeling abnormal events (such as circuit breaker time, recovery time, and anomaly type) based on these records; the training module uses an LSTM model to train a time-series prediction task, taking the past 30 minutes of transaction data as input and calculating a health index for the next 5 minutes; the decision module outputs a circuit breaker decision (yes / no) based on the health index; and a reinforcement learning model is used to optimize the circuit breaker threshold, where the reward function is to reduce the number of false circuit breakers and shorten the recovery time. After the dynamic circuit breaker decision model is trained, the decision module is applied during the circuit breaker and recovery processes.

[0051] An application example of the trained dynamic circuit breaker decision model is as follows: The inputs to the dynamic circuit breaker decision model are: a transfer success rate of 85%, an error code frequency of 100 times / minute, and a real-time transfer rate of 1800 TPS. The output of the dynamic circuit breaker decision model is: triggering the circuit breaker, with a recovery rate of 5% / minute.

[0052] This embodiment overcomes the limitations of traditional fixed-threshold circuit breakers by employing a pre-trained dynamic circuit breaker decision model for intelligent analysis. This model understands the complex operating environment of resource transfer channels, including time-series fluctuations and the correlations between multi-dimensional data. Through model prediction, it achieves dynamic calculation of the health index and personalized setting of the circuit breaker threshold, solving the problem that fixed-threshold circuit breaker schemes in related technologies cannot adapt to business fluctuations. Only if the health index of the current resource transfer channel is lower than the corresponding dynamic circuit breaker threshold across multiple consecutive historical timeframes will the system determine that the current resource transfer channel needs to be circuit-broken, thus improving the accuracy of the circuit breaker decision.

[0053] In one exemplary embodiment, such as Figure 3 As shown, the resource transfer channel data includes transfer success rate, real-time transfer volume, and error code frequency. The server maintains the resource transfer channel data in the database, and the data acquisition module of the dynamic circuit breaker decision model can also obtain the resource transfer channel data from the database.

[0054] The transfer success rate refers to the percentage of resource transfer requests that are successfully completed within a certain time window. It is typically calculated by dividing the number of successfully completed resource transfers by the total number of resource transfer requests. For example, if there are 1000 resource transfer requests in a given time period, and 900 are successful, then the transfer success rate is 90%. The transfer success rate is an important health indicator, reflecting the reliability and efficiency of the resource transfer channel.

[0055] Real-time transfer rate (TPS) represents the number of resource transfers per second and is a key indicator for measuring the load capacity and processing speed of a resource transfer channel. A higher TPS indicates greater real-time throughput and stronger processing capabilities. For example, a real-time transfer rate of 1500 TPS means the system can handle 1500 resource transfer requests per second.

[0056] Error codes are codes returned by the system when it encounters problems while processing resource transfer requests. They identify different error types. Examples include error codes such as "96 - System Error" and "53 - Insufficient Resource Reserves." Error code frequency refers to the number of times an error code occurs per unit of time, used to quantify abnormal conditions in the resource transfer channel. An increased error code frequency suggests potential instability or malfunction in the channel. For example, an error code "96" occurring 60 times per minute indicates that 60 resource transfer attempts fail per minute due to system abnormalities.

[0057] In some embodiments, a pre-trained dynamic circuit breaker decision model is used to analyze resource transfer channel data across multiple historical timeframes to obtain a health index for the current resource transfer channel in each of the multiple historical timeframes, including:

[0058] Using a pre-trained dynamic circuit breaker decision model, the first ratio between the error code frequency and the maximum error code frequency is calculated; the second ratio between the real-time transfer volume and the maximum transaction volume is calculated; the first ratio is mirrored to obtain the first indicator value, and the second ratio is mirrored to obtain the second indicator value; the transfer success rate, the first indicator value, and the second indicator value are weighted and summed to obtain the health index of the current resource transfer channel.

[0059] The maximum error code frequency is derived from historical data statistics and represents the highest frequency of error codes that the resource transfer system can tolerate under normal operating conditions.

[0060] Maximum transaction volume refers to the maximum processing capacity of a resource transfer channel under high load and high intensity operation, as recorded in historical data. Maximum transaction volume serves as a reference point to measure whether real-time transfer volume is approaching or exceeding the channel's limit.

[0061] Mirroring is a mathematical operation used to convert the first ratio and the second ratio into a more intuitive indicator value reflecting the health status of the channel. Mirroring typically maps the ratio to a preset indicator range (e.g., 1 to 0). In this embodiment, the first ratio represents the ratio between the frequency of error codes occurring in the current resource transfer channel and the maximum error code frequency that the channel can historically withstand, i.e., the relative frequency. Mirroring the first ratio yields a first indicator value that essentially reflects the degree to which the channel is in a normal state. In this embodiment, the mirroring of the first ratio is achieved by calculating the difference between the value 1 and the first ratio. The value 1 can be understood as a completely healthy or error-free state. Subtracting the value of "error code frequency / maximum frequency," the closer the result is to 1, the lower the current error code frequency, the closer the channel is to a normal state, and the higher its health. Similarly, the second ratio represents the ratio between the real-time transaction load of the current resource transfer channel and its historical maximum processable transaction volume, intuitively reflecting the current load level of the channel. A second ratio close to 1 means the channel is operating close to full capacity, while a ratio below 1 indicates that the current load of the channel is below its maximum processing capacity. The second index value obtained by mirroring the second ratio essentially reflects the remaining capacity or processing margin of the channel, which can be understood as the degree to which the channel is underloaded. In this embodiment, the mirroring of the second ratio is achieved by calculating the difference between the value 1 and the second ratio. The closer the second index value is to 1, the lower the real-time transfer volume of the channel, and the less fully the channel's processing capacity is utilized compared to the maximum transaction volume; conversely, the closer the value is to 0, the closer the channel is to full load operation, and the very small processing margin.

[0062] Optionally, the server uses a pre-trained dynamic circuit breaker decision model to calculate a first ratio between the error code frequency and the maximum error code frequency, and a second ratio between the real-time transfer volume and the maximum transaction volume; calculates the difference between the value 1 and the first ratio to obtain a first indicator value, calculates the difference between the value 1 and the second ratio to obtain a second indicator value; and performs a weighted summation of the transfer success rate, the first indicator value, and the second indicator value to obtain the health index of the current resource transfer channel.

[0063] For example, the health index of the current resource transfer channel can be expressed as: Health Index = 0.5 * Transfer Success Rate + 0.3 * (1 - Error Code Frequency / Maximum Error Code Frequency) + 0.2 * (1 - Real-time Transfer Volume / Maximum Transaction Volume), where 0.5, 0.3, and 0.2 represent weights.

[0064] An example of calculating the health index is: Health Index = 0.5 * 0.87 + 0.3 * (1 - 60 / 500) + 0.2 * (1 - 1500 / 2000) = 0.435 + 0.228 + 0.05 = 0.713.

[0065] This embodiment integrates three key indicators—error code frequency, real-time transfer volume (TPS), and transfer success rate—to construct a health index. Specifically, the first ratio of error code frequency to the maximum error code frequency, and the second ratio of real-time transfer volume to the maximum transaction volume, are both transformed into positive indicators through mirroring; that is, the closer to 1, the healthier the channel. This multi-dimensional data fusion analysis method can more comprehensively and accurately assess the real-time health status of resource transfer channels, solving the problem of inaccurate assessment by a single indicator. Mirroring is used to inversely express the relative frequency of error codes and the load ratio of real-time transfer volume as health indicators, converting the negative impact of abnormal situations into a positive assessment. This facilitates unified measurement with indicators such as success rate. Finally, the health index is obtained through weighted summation. This inverse indicator expression method improves the intuitiveness and operability of health assessment, avoids misjudgments that may arise from single-indicator assessments, and ensures the accuracy of circuit breaker and recovery decisions.

[0066] In one exemplary embodiment, each of the multiple historical time ranges is taken as the current historical time range, and the following prediction operation is performed to obtain the dynamic circuit breaker threshold of the current resource transfer channel in each historical time range:

[0067] First, by using a pre-trained dynamic circuit breaker decision model, the periodicity of traffic corresponding to the current historical time range is determined based on the resource transfer channel data within the current historical time range.

[0068] Traffic periodicity refers to the cyclical pattern of transaction volume changes in resource transfers across different time periods. It reflects the seasonal fluctuations in resource transfer operations, such as the difference between "busy" and "off-peak" periods on weekdays. In this embodiment, traffic periodicity is used to characterize whether the current resource transfer channel is in a period of high resource transfer volume or low resource transfer volume within the current historical time frame. High resource transfer volume periods refer to the time periods during which the resource transfer system experiences high transaction volumes or a large number of concurrent requests. This typically occurs during peak business periods, such as specific hours in the morning or afternoon on weekdays, or during special business events such as holidays or promotional activities. Low resource transfer volume periods refer to the time periods during which the resource transfer system experiences lower transaction volumes, typically occurring during non-peak business periods, such as late at night, early morning, or quiet afternoons on weekdays. This period corresponds to the traditional "off-peak period."

[0069] Optionally, Figure 4 This is a decision-making diagram of another optional dynamic circuit breaker decision-making model according to an embodiment of this application, such as... Figure 4As shown, the server uses a pre-trained dynamic circuit breaker decision model to collect resource transfer channel data (including real-time transfer volume) within the current historical time range and compares it with a preset maximum capacity threshold (e.g., 2000 TPS). If the transaction volume approaches or exceeds the maximum capacity (e.g., reaches 1800 TPS or more), it is determined that the current historical time range's traffic periodicity indicates that the current resource transfer channel is in a period of intensive resource transfer within that time range. Similarly, if the real-time transfer volume within the current historical time range is less than the minimum capacity threshold (e.g., 500 TPS), it is determined that the current historical time range's traffic periodicity indicates that the current resource transfer channel is in a period of idle resource transfer within that time range.

[0070] In some embodiments, the server can also use a pre-trained dynamic circuit breaker decision model to predict the future transaction volume of the current resource transfer channel based on the resource transfer channel data within the current historical time range. If the difference between the future transaction volume and the real-time transfer volume within the current historical time range is greater than a first difference threshold, it indicates that the predicted future transaction volume will increase sharply, and the periodicity of the traffic corresponding to the current historical time range indicates that the current resource transfer channel is in a period of intensive resource transfer within the current historical time range. Similarly, if the absolute value of the difference between the future transaction volume and the real-time transfer volume within the current historical time range is greater than a second difference threshold, it indicates that the periodicity of the traffic corresponding to the current historical time range indicates that the current resource transfer channel is in a period of idle resource transfer within the current historical time range.

[0071] Here, future transaction volume refers to the prediction of the expected transaction request volume of the resource transfer channel within a specified time range (e.g., the next 5 minutes) after the current historical time range. In this embodiment, the dynamic circuit breaker decision model is a time series prediction model such as an LSTM model. By calling a pre-trained dynamic circuit breaker decision model, based on the resource transfer channel data of the current historical time range, the future transaction volume of the current resource transfer channel is predicted, providing a basis for the dynamic adjustment of the circuit breaker threshold.

[0072] 2. When the current historical time range corresponds to a period of intensive resource transfer, the dynamic circuit breaker threshold for the current historical time range is determined as the first circuit breaker threshold. When the current historical time range corresponds to a period of slack resource transfer, the dynamic circuit breaker threshold for the current historical time range is determined as the second circuit breaker threshold. The second circuit breaker threshold is greater than the first circuit breaker threshold.

[0073] The first circuit breaker threshold is a dynamic threshold set when the current resource transfer channel is expected or detected to be in a period of intensive resource transfer. This threshold considers predicted future transaction volume and real-time transaction data, aiming to ensure basic stability and availability even during peak business periods. It avoids misjudging channel anomalies due to normal business pressure (such as response delays caused by high concurrency), thereby reducing false circuit breakers. For example, if future transaction volume surges to 2000 TPS, determining that the current historical timeframe falls within a period of intensive resource transfer, the success rate circuit breaker threshold may be lowered from 90% to even lower (e.g., 85%) during this period. Circuit breaking is only triggered when the actual success rate consistently falls below 85%, reducing the possibility of false circuit breakers and ensuring effective handling of sudden high concurrency situations.

[0074] The second circuit breaker threshold is a dynamic threshold applied when the current resource transfer channel determines that it is in a resource transfer idle period. The second circuit breaker threshold is higher than the first, meaning that during periods of low trading volume, the response to anomalies will be more sensitive, allowing for timely detection and handling of potential problems. For example, if the future trading volume is only 300 TPS, and the current historical time range is determined to be in a resource transfer idle period, the circuit breaker success rate threshold may be adjusted from 90% to a higher level (e.g., 92%). If the success rate drops to 90%, the circuit breaker will be triggered immediately, quickly isolating potential problems. This ensures that even in sparse trading conditions, potential errors or anomalies can be quickly identified, allowing for appropriate circuit breaker measures to be taken, reducing the complexity of subsequent recovery and the need for manual intervention.

[0075] This embodiment determines the periodicity of traffic in the current resource transfer channel within different historical time ranges. It can intelligently set circuit breaker thresholds based on the characteristics of peak and off-peak resource transfer periods. During peak resource transfer periods, a lower first circuit breaker threshold is set to increase tolerance for high-concurrency transactions and avoid premature circuit breaking. During off-peak resource transfer periods, a higher second circuit breaker threshold is used to enhance sensitivity to anomalies and errors, enabling timely detection and handling of problems. This method of dynamically adjusting the circuit breaker threshold according to business needs solves the problem of fixed threshold strategies being inflexible in different scenarios, improving system stability and response efficiency.

[0076] In an exemplary embodiment, existing circuit breaker schemes typically employ fixed-interval activation during the recovery phase, which may lead to recovery being too fast or too slow, increasing system risk. Therefore, to address this issue, this embodiment uses a pre-trained dynamic circuit breaker decision model to analyze resource transfer channel data across multiple historical timeframes to derive a predicted recovery strategy, including:

[0077] Using a pre-trained dynamic circuit breaker decision model, real-time cases are generated based on resource transfer channel data across multiple historical timeframes. Target historical cases matching the real-time cases are then searched in a pre-built knowledge base. The knowledge base stores multiple historical cases; each historical case includes anomalies in the resource transfer channel and corresponding historical circuit breaker strategies. The pre-trained dynamic circuit breaker decision model determines the historical circuit breaker strategy corresponding to the target historical case as the predictive recovery strategy.

[0078] Real-time cases refer to current operational status cases generated based on real-time monitoring data of the current resource transfer channel, including but not limited to transfer success rate, error code frequency, and real-time transfer volume (TPS). Real-time cases are the input to the dynamic circuit breaker decision model. The model compares cases with those stored in the historical knowledge base, automatically matches the most similar historical cases, and thus predicts the most appropriate circuit breaker and recovery strategy.

[0079] A knowledge base is a database or information repository that stores multiple historical cases. Each historical case contains human experience in handling specific abnormal events in resource transfer channels, such as circuit breaker timing, recovery strategies, and exception types. The knowledge base is an important data source for the system to make intelligent decisions. By learning from historical cases, the dynamic circuit breaker decision model can automatically identify the pattern of the current abnormal event, call the most relevant circuit breaker strategy, and realize the intelligent application of experience.

[0080] Each historical case in the knowledge base includes anomalies in the resource transfer channel and corresponding historical circuit breaker strategies. The historical circuit breaker strategies refer to the manual circuit breaker operations and recovery plans implemented for each anomaly in historical cases, including circuit breaker threshold settings, initial traffic recovery ratios after circuit breaker intervention, and the execution frequency of the recovery strategy.

[0081] The construction of the knowledge base includes the following steps: collecting historical manual circuit breaker records (including circuit breaker time, recovery time, anomaly type (such as system upgrade, network jitter, etc.)) and manual handling strategies (such as waiting 10 minutes after circuit breaker before recovery); tagging the historical manual handling records to annotate abnormal events, clearly indicating the circuit breaker time, recovery time, and anomaly type of each event to form a structured event description; using natural language processing (NLP) technology to perform semantic analysis on the descriptions of abnormal events, extracting key features, analyzing the correlation between abnormal events and circuit breaker strategies, and constructing a mapping relationship between anomaly types and circuit breaker strategies; integrating the processed historical data into the knowledge base, where each case includes not only historical transaction data and error code data, but also related circuit breaker time, recovery strategies, and manual handling experience.

[0082] Once the knowledge base is built, when a similar abnormal pattern is detected, the server automatically retrieves historical handling strategies from the knowledge base. For example, if the frequency of error code "96" increases and transaction volume drops sharply, matching the "system upgrade" scenario in the historical records, the server automatically extends the recovery time to 10 minutes after the channel circuit breaker is broken, avoiding frequent active probes that could cause channel pressure.

[0083] Optionally, such as Figure 4 As shown, the server integrates resource transfer channel data from multiple historical time ranges using a pre-trained dynamic circuit breaker decision model to form real-time cases describing the current state of the resource transfer channels. In a pre-built knowledge base, a similarity algorithm, such as a classification model based on cosine similarity or machine learning, is used to find the target historical case that best matches the real-time case. The server then extracts the corresponding historical circuit breaker strategy from the matched target historical case using the pre-trained dynamic circuit breaker decision model, and determines the extracted historical circuit breaker strategy as the predicted recovery strategy. The dynamic circuit breaker decision model automatically executes circuit breaker and traffic recovery operations according to preset rules and strategies, such as reducing the real-time transfer volume to 1% and gradually restoring traffic according to the predicted recovery strategy.

[0084] In this embodiment, a knowledge base is constructed that stores multiple historical cases. Each case not only records abnormal events in the resource transfer channel but also includes the corresponding manual circuit breaker strategy. When a real-time case is generated, the dynamic circuit breaker decision model automatically searches the knowledge base for the target historical case that best matches the real-time case. By analyzing the similarity of abnormal events, it invokes the manual handling strategy from the historical case. This method overcomes the problem in traditional solutions where the recovery phase typically uses fixed-interval liveness detection, which cannot adapt to new scenarios, leading to recovery that is too fast or too slow. It achieves intelligent application of experience and improves the accuracy of anomaly handling.

[0085] In one exemplary embodiment, at the moment indicated by the recovery timing, restoring the current resource transfer channel according to the traffic recovery rate includes:

[0086] After the circuit breaker duration of the current resource transfer channel reaches the preset duration, the second flow is released to the current resource transfer channel to perform a liveness test on the current resource transfer channel and monitor the first test information returned by the current resource transfer channel; the second flow is greater than the first flow; if the first test information meets the preset recovery conditions, the current resource transfer channel is restored according to the flow recovery rate.

[0087] The circuit breaker duration refers to the duration of the resource transfer channel under circuit breaker status, while the preset duration refers to the timing of the recovery after the current resource transfer channel is circuit breaker broken. When the actual circuit breaker duration reaches or exceeds the preset duration, the recovery mechanism will be activated to avoid prolonged service interruption.

[0088] The second flow rate is the flow rate used to conduct preliminary activation tests on the resource transfer channel after the circuit breaker duration reaches the preset duration. The second flow rate is higher than the first flow rate.

[0089] Active recovery testing refers to sending a small number of requests (e.g., 10 requests / second) to the current resource transfer channel after it has been disrupted, using a second set of traffic as a baseline. This is done to check if the current state of the resource transfer channel is suitable for gradual traffic recovery and to assess its ability to return to normal service. Active recovery testing also monitors the channel's immediate response, specifically the first test information returned by the current resource transfer channel. This first test information refers to the information returned by the current resource transfer channel during the active recovery test, used to assess its health status, including but not limited to success rate, response time, and error code distribution. The first test information is a crucial basis for determining whether the current resource transfer channel is ready for traffic recovery.

[0090] Preset recovery conditions are a series of evaluation criteria set in the intelligent recovery process of the resource transfer channel to ensure that the channel can safely and stably return to normal service after the circuit breaker is lifted. These typically include, but are not limited to, success rate thresholds, response time, error code frequency, load balancing, and stability testing.

[0091] Optionally, the server continuously monitors the circuit breaker status of the current resource transfer channel. Once the circuit breaker duration is detected to have reached a preset threshold, a recovery preparation process is immediately triggered. First, a second flow exceeding the first flow is released to the circuit breaker channel for a liveness test to assess the channel's immediate state before recovery. During the liveness test, the server automatically collects the first test information returned by the channel, including but not limited to transfer success rate, response time, and error code distribution. Through real-time data analysis, it determines whether the channel meets the preset recovery conditions. If not, the server uses an exponential backoff algorithm to dynamically adjust the liveness test interval (e.g., 30s → 5min → 10min) and repeats subsequent tests until the channel stabilizes. If the conditions are met, the server will gradually increase the channel's flow based on the dynamically generated flow recovery rate (e.g., 5% / minute) from the dynamic circuit breaker decision model until normal service levels are restored.

[0092] In this embodiment, after the resource transfer channel is interrupted, instead of immediately restoring all traffic, a second level of traffic, higher than the first level, is released first to conduct a probe test. This intelligently assesses the channel's immediate response status and recovery potential. By monitoring the first test information obtained from the probe test, the stability and health status of the channel can be judged based on real-time feedback. This ensures that traffic return only begins when the channel has truly recovered to an acceptable state, preventing the risks of blind recovery. It overcomes the channel overload risk that may be caused by the fixed recovery strategy in traditional solutions and avoids the problem of re-interruption caused by a sudden surge of traffic during the recovery process.

[0093] In an exemplary embodiment, the first test information includes the transfer success rate and error code frequency. When the first test information meets preset recovery conditions, the current resource transfer channel is restored according to the traffic recovery rate, including:

[0094] If the transfer success rate is greater than the preset success rate threshold and the error code frequency is less than the preset error frequency threshold, the current resource transfer channel is restored according to the traffic recovery rate.

[0095] In this embodiment, the preset recovery conditions are a transfer success rate greater than a preset success rate threshold and an error code frequency less than a preset error frequency threshold. The transfer success rate is a core indicator for measuring the performance of a resource transfer channel service. Setting a high preset success rate threshold (e.g., 95%) means that when restoring the channel, the system must first confirm that the channel can successfully complete the resource transfer task with a high probability, which directly relates to the channel's ability to restore normal service. Only when the success rate exceeds the preset threshold can the channel be considered to have recovered to a sufficiently safe and reliable level in terms of processing capacity, avoiding a decline in user service experience due to premature recovery. The error code frequency reflects the frequency with which the channel encounters abnormal situations during resource transfer. When the error code frequency is higher than a preset error frequency threshold (e.g., 5 times / minute), it indicates that there are unresolved potential problems in the channel, which may stem from system failures, network latency, or other technical obstacles. Using the error code frequency as one of the recovery conditions helps the system identify whether the channel has truly recovered from an abnormal state, avoiding triggering the circuit breaker again due to frequent errors, and ensuring service continuity and stability. Setting these two conditions as necessary conditions for recovery effectively finds a balance between efficiency and security. Premature or overly aggressive recovery may lead to overload due to incomplete channel restoration, while overly conservative recovery will prolong service interruption time, affecting user experience and business continuity. By dynamically monitoring success rate and error code frequency, the system can intelligently determine the timing of recovery, ensuring that the channel is restored to normal service status as quickly as possible while meeting security requirements, thereby improving overall operational efficiency and user satisfaction.

[0096] In this embodiment, based on two key indicators monitored in real time, namely "transfer success rate is greater than the preset success rate threshold" and "error code frequency is less than the preset error frequency threshold", the system can intelligently determine whether the channel is ready for recovery, thus avoiding the risk of premature or delayed recovery caused by channel recovery relying on a fixed duration or manual judgment in related technologies.

[0097] In one exemplary embodiment, restoring the current resource transfer channel according to the traffic recovery rate includes:

[0098] Based on the second flow rate, the flow rate of the current resource transfer channel is gradually increased according to the flow recovery rate until the flow rate of the current resource transfer channel is restored to the maximum flow rate, thus obtaining the restored current resource transfer channel.

[0099] The second flow rate refers to the initial flow rate used to test the channel recovery status after the channel is tripped. It is generally much lower than the maximum flow rate to facilitate a safe assessment of the current health status of the channel.

[0100] Maximum traffic volume refers to the upper limit of traffic volume of the resource transfer channel under normal operating conditions, that is, the traffic capacity that the channel can stably and efficiently handle. Maximum traffic volume is used to indicate the highest number of transaction requests or data transmissions that the resource transfer channel can handle under normal operating conditions; it is an upper limit indicator of the channel's capacity. The goal of the recovery process is to gradually restore the channel traffic to maximum traffic volume, ensuring the complete restoration of channel service. In this embodiment, for example, maximum traffic volume is 100%. When the channel is at 100% traffic volume, it means that it is operating at its maximum data processing capacity or transaction request volume.

[0101] Optionally, if the circuit breaker duration of the current resource transfer channel reaches the preset duration indicated by the predicted recovery strategy, and the current resource transfer channel meets the preset recovery conditions, the server gradually increases the flow of the current resource transfer channel according to the flow recovery rate, for example, increasing the flow by 5% per minute. At each stage of the flow increase, the server needs to continuously monitor various indicators of the current resource transfer channel to ensure the stability and security of the recovery process. If any anomaly is detected (such as a sudden drop in transfer success rate or a sudden increase in error codes), the flow increase is immediately paused, and the flow may even revert to the previous stable state. Subsequently, the status of the current resource transfer channel is reassessed until the maximum flow (100%) is reached, and the current resource transfer channel is restored.

[0102] In this embodiment, based on the second flow rate, the flow rate of the current resource transfer channel is gradually increased according to the flow recovery rate. By gradually increasing the flow rate instead of restoring it to the normal level all at once, the circuit breaker mechanism can be avoided from being triggered again due to the channel suddenly being subjected to a large number of requests. This strategy is like giving the channel a "buffer period" so that it has time to gradually adapt to the increased load, thereby ensuring the smoothness of the recovery process and the overall stability of the system.

[0103] In one exemplary embodiment, based on the second flow rate, the flow rate of the current resource transfer channel is gradually increased according to the flow recovery rate until the flow rate of the current resource transfer channel recovers to the maximum flow rate, thereby obtaining the recovered current resource transfer channel, including:

[0104] First, based on the second flow rate, increase the flow rate of the current resource transfer channel according to the flow recovery rate to obtain the third flow rate, and monitor the second test information of the current resource transfer channel under the third flow rate within the specified monitoring period.

[0105] The third flow rate refers to the flow level increased based on the second flow rate, according to the flow recovery rate determined by the dynamic circuit breaker decision model. During channel recovery, the third flow rate is the next test point in the recovery process. It can be understood that the third flow rate is greater than the second flow rate.

[0106] The specified monitoring duration refers to the period during which key metrics (such as success rate, response time, and error code distribution) of the channel are continuously monitored after the traffic on the resource transfer channel increases to level three. The specified monitoring duration aims to ensure the channel remains stable under high traffic conditions, providing a basis for subsequent recovery decisions.

[0107] The second test information refers to real-time test data of the resource transfer channel under third-level traffic, including but not limited to key indicators such as transfer success rate, average response time, and error code distribution. The second test information is used to evaluate the channel's performance and stability at higher traffic levels (i.e., third-level traffic). "First test information" and "second test information" correspond to test datasets of the current resource transfer channel at different recovery stages, with the main difference being the test time and traffic level. Specifically, the first test information is the real-time transaction data and error code data obtained during the initial probing test of the channel at a very low traffic level (i.e., second-level traffic, typically 1% of the maximum traffic) in the initial stage after the channel circuit breaker is triggered. The first test information is used to assess the basic recovery and health status of the channel under minimum pressure, determining whether to begin gradually increasing traffic; it is the first step in the recovery process. The second test information is more comprehensive real-time transaction data and channel performance indicators collected within a specified monitoring period as the channel traffic gradually increases from second-level to third-level. The second test information is used to further evaluate the channel's stability and carrying capacity under higher traffic pressure, determining whether to continue increasing traffic to a higher level or reduce traffic to a lower level.

[0108] Optionally, if the current resource transfer channel activation test is successful, the server will adjust the second traffic level of the current resource transfer channel to the third traffic level according to the traffic recovery rate of the next stage; after the traffic is increased to the third traffic level, the server will start real-time monitoring for a specified monitoring duration (e.g., 2 minutes) to continuously collect second test information, including key indicators such as success rate, average response time, and error code distribution.

[0109] Second, if the second test information meets the preset recovery conditions, the third traffic is taken as the new second traffic. Based on the new second traffic, the process of increasing the traffic of the current resource transfer channel according to the traffic recovery rate continues until the traffic of the current resource transfer channel is restored to the maximum traffic, thus obtaining the restored current resource transfer channel.

[0110] In this embodiment, the preset recovery conditions are a series of evaluation criteria set in the intelligent recovery process of the resource transfer channel to ensure that the channel can safely and stably return to normal service after the circuit breaker is lifted. For example, the preset recovery conditions may be that the transfer success rate is greater than a preset success rate threshold and the error code frequency is less than a preset error frequency threshold.

[0111] Optionally, if the second test information meets the preset recovery conditions, the server sets the current third traffic (the increased traffic) as the new second traffic and prepares for the next round of traffic enhancement testing. During the next round of traffic enhancement testing, based on the new second traffic, the traffic of the current resource transfer channel is increased according to the traffic recovery rate to obtain the third traffic. The server monitors the second test information of the current resource transfer channel under the third traffic within a specified monitoring period. If the second test information meets the preset recovery conditions, the step of setting the current third traffic (the increased traffic) as the new second traffic is repeated until the traffic of the current resource transfer channel recovers to the maximum traffic, thus obtaining the recovered current resource transfer channel.

[0112] For example, a traffic boosting test could be performed as follows: In the first round of traffic boosting testing, the traffic of the current resource transfer channel is increased from the first flow (e.g., 1%) to the second flow (e.g., 6%), and the first test information of the current resource transfer channel under the second flow (e.g., 6%) is monitored within a specified monitoring period (e.g., 2 minutes). If the first test information meets the preset recovery conditions, the second round of traffic boosting testing begins. In the second round of traffic boosting testing, the traffic of the current resource transfer channel is increased from the second flow (e.g., 6%) to the third flow (e.g., 11%), and the second test information of the current resource transfer channel under the third flow (e.g., 11%) is monitored within a specified monitoring period (e.g., 2 minutes). If the second test information meets the preset recovery conditions, the third round of traffic boosting testing begins. In the third round of traffic boosting test, the traffic of the current resource transfer channel is increased from the third traffic (e.g., 11%) to the fourth traffic (e.g., 16%), and the third test information of the current resource transfer channel under the fourth traffic (e.g., 16%) is monitored within a specified monitoring period (e.g., 2 minutes). If the third test information meets the preset recovery conditions, the fourth round of traffic boosting test is entered, and so on, until the traffic of the current resource transfer channel is restored to the maximum traffic (e.g., 100%), and the restored current resource transfer channel is obtained.

[0113] This embodiment, by gradually increasing traffic, allows for more detailed monitoring of the channel recovery process, ensuring that each traffic increase is carried out under the premise that the channel can handle it stably. This avoids the risk of recurring anomalies that may result from directly restoring to high traffic. In each round of traffic increase testing, the second test information is checked within a specified monitoring period to see if it meets the preset recovery conditions. Only when the second test information meets the preset recovery conditions will the next round of traffic increase testing be conducted. This mechanism can promptly detect and handle potential anomalies, reduce the impact on business caused by erroneous recovery, and effectively control the risks in the recovery process.

[0114] In one exemplary embodiment, the above-described resource transfer channel control method further includes:

[0115] If the second test information does not meet the preset recovery conditions, the traffic of the current resource transfer channel will be reverted to the traffic in the previous traffic increase phase, and the step of performing a liveness test on the current resource transfer channel will be returned to continue until the traffic of the current resource transfer channel is restored to the maximum traffic, thus obtaining the restored current resource transfer channel.

[0116] The "traffic in the previous traffic increase phase" refers to the traffic level during the most recent traffic increase test, i.e., the traffic state the channel was in before the current test failed. The traffic callback mechanism ensures that the channel can quickly revert to the previous stable state when an anomaly occurs, preventing further deterioration of the problem. It also provides a safe starting point for subsequent recovery, effectively achieving the safe and stable recovery of the resource transfer channel.

[0117] Optionally, during the current traffic boost test, if the second test information monitored during the current traffic boost test does not meet the preset recovery conditions, the server will revert the traffic of the current resource transfer channel to the traffic in the previous traffic increase phase, and use the traffic in the previous traffic increase phase as the starting point for the liveness detection test to continue the liveness detection test. The second test information will be collected again based on the reverted traffic, and performance evaluation will be performed until the traffic of the current resource transfer channel is restored to the maximum traffic, thus obtaining the restored current resource transfer channel.

[0118] For example, if an anomaly is detected in the third round of traffic boosting test (currently the traffic of the resource transfer channel is 16%), the traffic is reduced to the previous stable stage (e.g., from 16% to 11%), the liveness test is restarted, and the channel status is evaluated.

[0119] In this embodiment, when the second test information monitored in real time does not meet the preset recovery conditions, the traffic of the current resource transfer channel is automatically reverted to the traffic level of the previous stage. The innovation of this mechanism is that it is not based on a fixed threshold or manual command, but is adjusted according to real-time data. Compared with the traditional fixed threshold, it has stronger adaptability and robustness, can effectively cope with business fluctuations and sudden anomalies, reduce the risk of false recovery, and avoid prolonging the recovery time due to excessive caution.

[0120] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0121] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0122] According to another aspect of the embodiments of this application, a resource transfer channel control device is also provided. This resource transfer channel control device can be used to implement the resource transfer channel control method provided in the above embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0123] Figure 5 This is a structural block diagram of an optional resource transfer channel control device according to an embodiment of this application, such as... Figure 5 As shown, the resource transfer channel control device includes:

[0124] The acquisition module 502 is used to acquire resource transfer channel data of the current resource transfer channel within multiple consecutive historical time ranges.

[0125] The dynamic decision module 504 is used to analyze resource transfer channel data across multiple historical time ranges using a pre-trained dynamic circuit breaker decision model to obtain circuit breaker decision results and predicted recovery strategies. The predicted recovery strategy specifies the recovery timing and traffic recovery rate of the current resource transfer channel. The recovery timing refers to the timing of restoring the current resource transfer channel after a preset time after the current resource transfer channel is circuit breaker is broken. The traffic recovery rate refers to the rate at which traffic increases during the restoration of the current resource transfer channel.

[0126] The channel control module 506 is used to, when the circuit breaker decision result indicates that the current resource transfer channel is to be broken, to reduce the flow of the current resource transfer channel to a first flow, and to restore the current resource transfer channel at the time indicated by the recovery timing, according to the flow recovery rate.

[0127] In an exemplary embodiment, the dynamic decision module 504 is further configured to analyze resource transfer channel data across multiple historical time ranges using a pre-trained dynamic circuit breaker decision model, to obtain the health index of the current resource transfer channel in each historical time range and the dynamic circuit breaker threshold of the current resource transfer channel in each historical time range; and, if the health index in each historical time range is less than the corresponding dynamic circuit breaker threshold, to output a circuit breaker decision result representing the circuit breaker decision of suspending the current resource transfer channel through the dynamic circuit breaker decision model.

[0128] In an exemplary embodiment, the resource transfer channel data includes transfer success rate, real-time transfer volume, and error code frequency; the dynamic decision module 504 is further configured to calculate a first ratio between the error code frequency and the maximum error code frequency using a pre-trained dynamic circuit breaker decision model; calculate a second ratio between the real-time transfer volume and the maximum transaction volume; perform a mirror transformation on the first ratio to obtain a first indicator value; perform a mirror transformation on the second ratio to obtain a second indicator value; and perform a weighted summation of the transfer success rate, the first indicator value, and the second indicator value to obtain the health index of the current resource transfer channel.

[0129] In an exemplary embodiment, the dynamic decision module 504 is further configured to take each of the multiple historical time ranges as the current historical time range and perform the following prediction operation to obtain the dynamic circuit breaker threshold of the current resource transfer channel in each historical time range: using a pre-trained dynamic circuit breaker decision model, based on the resource transfer channel data in the current historical time range, determine the traffic periodicity corresponding to the current historical time range; if the traffic periodicity corresponding to the current historical time range indicates that the current historical time range is in a period of intensive resource transfer, determine the dynamic circuit breaker threshold of the current historical time range as a first circuit breaker threshold; if the traffic periodicity corresponding to the current historical time range indicates that the current historical time range is in a period of idle resource transfer, determine the dynamic circuit breaker threshold of the current historical time range as a second circuit breaker threshold, wherein the second circuit breaker threshold is greater than the first circuit breaker threshold.

[0130] In an exemplary embodiment, the dynamic decision module 504 is further configured to generate real-time cases based on resource transfer channel data within multiple historical time ranges using a pre-trained dynamic circuit breaker decision model, and search for target historical cases matching the real-time cases in a pre-built knowledge base; the knowledge base stores multiple historical cases; each historical case includes an abnormal event of the resource transfer channel and the corresponding historical circuit breaker strategy; the pre-trained dynamic circuit breaker decision model determines the historical circuit breaker strategy corresponding to the target historical case as the predictive recovery strategy.

[0131] In an exemplary embodiment, the channel control module 506 is further configured to release a second flow to the current resource transfer channel after the circuit breaker duration of the current resource transfer channel reaches a preset duration, perform a liveness test on the current resource transfer channel, and monitor the first test information returned by the current resource transfer channel; the second flow is greater than the first flow; and if the first test information meets the preset recovery conditions, restore the current resource transfer channel according to the flow recovery rate.

[0132] In one exemplary embodiment, the first test information includes the transfer success rate and the error code frequency; the channel control module 506 is further configured to restore the current resource transfer channel according to the traffic recovery rate when the transfer success rate is greater than a preset success rate threshold and the error code frequency is less than a preset error frequency threshold.

[0133] In an exemplary embodiment, the channel control module 506 is further configured to, based on the second flow rate, gradually increase the flow rate of the current resource transfer channel according to the flow recovery rate until the flow rate of the current resource transfer channel is restored to the maximum flow rate, thereby obtaining the restored current resource transfer channel.

[0134] In an exemplary embodiment, the channel control module 506 is further configured to increase the flow of the current resource transfer channel according to the flow recovery rate based on the second flow to obtain a third flow, and monitor the second test information of the current resource transfer channel under the third flow within a specified monitoring period; if the second test information meets the preset recovery conditions, the third flow is used as the new second flow, and based on the new second flow, the step of increasing the flow of the current resource transfer channel according to the flow recovery rate based on the second flow is returned to continue execution until the flow of the current resource transfer channel is restored to the maximum flow, thereby obtaining the restored current resource transfer channel.

[0135] In an exemplary embodiment, the channel control module 506 is further configured to, if the second test information does not meet the preset recovery conditions, revert the traffic of the current resource transfer channel to the traffic in the previous traffic increase phase, and return to the step of performing a liveness test on the current resource transfer channel until the traffic of the current resource transfer channel is restored to the maximum traffic, thereby obtaining the restored current resource transfer channel.

[0136] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0137] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein the program executes the steps in any of the above method embodiments when it is run.

[0138] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0139] According to another aspect of the embodiments of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is configured to perform the steps of any of the method embodiments described above via the computer program. In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0140] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0141] According to another aspect of the embodiments of this application, a computer program product is also provided, comprising a computer program / instructions containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit 601, it performs various functions provided in the embodiments of this application. The sequence numbers of the embodiments of this application above are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0142] Figure 6 A schematic block diagram of a computer system architecture for implementing embodiments of the present application is shown. Figure 6 As shown, the computer system 600 includes a CPU (Central Processing Unit) 601, which can perform various appropriate actions and processes based on programs stored in ROM 602 or programs loaded into RAM 603 from storage section 608. Random access memory 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via bus 604. An I / O (Input / Output) interface 605 is also connected to bus 604.

[0143] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including CRT (Cathode Ray Tube), LCD (Liquid Crystal Display), and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card, such as a LAN card or modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0144] Specifically, according to embodiments of this application, the processes described in the various method flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit 601, it performs various functions defined in the system of this application.

[0145] It should be noted that, Figure 6 The computer system 600 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0146] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0147] The above are merely preferred embodiments of this application and are not intended to limit 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 principles of this application should be included within the protection scope of this application.

Claims

1. A method for controlling resource transfer channels, characterized in that, include: Obtain resource transfer channel data for the current resource transfer channel across multiple consecutive historical time ranges; By analyzing the resource transfer channel data within the multiple historical time ranges using a pre-trained dynamic circuit breaker decision model, circuit breaker decision results and predicted recovery strategies are obtained. The predicted recovery strategy is used to specify the recovery timing and traffic recovery rate of the current resource transfer channel. The recovery timing refers to the timing of restoring the current resource transfer channel after a preset time after the current resource transfer channel is circuit breaker-broken. The traffic recovery rate refers to the rate at which traffic increases during the process of restoring the current resource transfer channel; If the circuit breaker decision indicates that the current resource transfer channel is to be breakered, the current resource transfer channel is breakered, the flow of the current resource transfer channel is reduced to a first flow, and the current resource transfer channel is restored at the time indicated by the recovery timing, according to the flow recovery rate.

2. The method according to claim 1, characterized in that, The pre-trained dynamic circuit breaker decision model analyzes resource transfer channel data across multiple historical timeframes to obtain circuit breaker decision results, including: By analyzing the resource transfer channel data in the multiple historical time ranges through a pre-trained dynamic circuit breaker decision model, the health index of the current resource transfer channel in each of the multiple historical time ranges and the dynamic circuit breaker threshold of the current resource transfer channel in each of the multiple historical time ranges are obtained. If the health index is less than the corresponding dynamic circuit breaker threshold in each historical time range, the dynamic circuit breaker decision model outputs a circuit breaker decision result that represents the circuit breaker decision to suspend the current resource transfer channel.

3. The method according to claim 2, characterized in that, The resource transfer channel data includes transfer success rate, real-time transfer volume, and error code frequency; the pre-trained dynamic circuit breaker decision model analyzes the resource transfer channel data across multiple historical time ranges to obtain the health index of the current resource transfer channel in each of the multiple historical time ranges, including: Using a pre-trained dynamic circuit breaker decision model, a first ratio between the error code frequency and the maximum error code frequency is calculated; a second ratio between the real-time transfer volume and the maximum transaction volume is calculated; the first ratio is mirrored to obtain a first indicator value, and the second ratio is mirrored to obtain a second indicator value; the transfer success rate, the first indicator value, and the second indicator value are weighted and summed to obtain the health index of the current resource transfer channel.

4. The method according to claim 2, characterized in that, The pre-trained dynamic circuit breaker decision model analyzes resource transfer channel data across multiple historical timeframes to obtain the dynamic circuit breaker threshold for the current resource transfer channel in each historical timeframe, including: Using each of the multiple historical time ranges as the current historical time range, perform the following prediction operation to obtain the dynamic circuit breaker threshold of the current resource transfer channel in each historical time range: By using a pre-trained dynamic circuit breaker decision model, the periodicity of traffic corresponding to the current historical time range is determined based on the resource transfer channel data within the current historical time range. If the periodicity of the traffic corresponding to the current historical time range indicates that the current historical time range is in a period of intensive resource transfer, then the dynamic circuit breaker threshold of the current historical time range is determined as the first circuit breaker threshold. When the periodicity of the traffic corresponding to the current historical time range indicates that the current historical time range is in a period of idle resource transfer, the dynamic circuit breaker threshold of the current historical time range is determined as the second circuit breaker threshold, wherein the second circuit breaker threshold is greater than the first circuit breaker threshold.

5. The method according to claim 2, characterized in that, The pre-trained dynamic circuit breaker decision model analyzes resource transfer channel data across multiple historical timeframes to obtain a predicted recovery strategy, including: The pre-trained dynamic circuit breaker decision model generates real-time cases based on resource transfer channel data within multiple historical time ranges, and searches for target historical cases that match the real-time cases in a pre-built knowledge base; the knowledge base stores multiple historical cases; each historical case includes abnormal events of resource transfer channels and corresponding historical circuit breaker strategies. By using a pre-trained dynamic circuit breaker decision model, the historical circuit breaker strategies corresponding to the target historical cases are determined as the predicted recovery strategies.

6. The method according to claim 1, characterized in that, Restoring the current resource transfer channel at the time indicated by the restoration timing, according to the traffic restoration rate, includes: After the circuit breaker duration of the current resource transfer channel reaches the preset duration, a second flow is released to the current resource transfer channel to perform a liveness test on the current resource transfer channel and monitor the first test information returned by the current resource transfer channel; the second flow is greater than the first flow. If the first test information meets the preset recovery conditions, the current resource transfer channel is restored according to the traffic recovery rate.

7. The method according to claim 6, characterized in that, The first test information includes the transfer success rate and error code frequency; the step of restoring the current resource transfer channel according to the traffic recovery rate when the first test information meets the preset recovery conditions includes: If the transfer success rate is greater than a preset success rate threshold and the error code frequency is less than a preset error frequency threshold, the current resource transfer channel is restored according to the traffic recovery rate.

8. The method according to claim 6 or 7, characterized in that, Restoring the current resource transfer channel according to the traffic recovery rate includes: Based on the second flow rate, the flow rate of the current resource transfer channel is gradually increased according to the flow recovery rate until the flow rate of the current resource transfer channel is restored to the maximum flow rate, thus obtaining the restored current resource transfer channel.

9. The method according to claim 8, characterized in that, Based on the second flow rate, the flow rate of the current resource transfer channel is gradually increased according to the flow recovery rate until the flow rate of the current resource transfer channel recovers to the maximum flow rate, thereby obtaining the recovered current resource transfer channel, including: Based on the second flow rate, the flow rate of the current resource transfer channel is increased according to the flow recovery rate to obtain the third flow rate, and the second test information of the current resource transfer channel under the third flow rate is monitored within a specified monitoring period. If the second test information meets the preset recovery conditions, the third traffic is used as the new second traffic, and based on the new second traffic, the step of increasing the traffic of the current resource transfer channel according to the traffic recovery rate based on the second traffic is returned to continue execution until the traffic of the current resource transfer channel is restored to the maximum traffic, and the restored current resource transfer channel is obtained.

10. The method according to claim 9, characterized in that, The method further includes: If the second test information does not meet the preset recovery conditions, the traffic of the current resource transfer channel is reverted to the traffic in the previous traffic increase phase, and the step of performing a liveness test on the current resource transfer channel is returned to continue execution until the traffic of the current resource transfer channel is restored to the maximum traffic, thus obtaining the restored current resource transfer channel.

11. A resource transfer channel control device, characterized in that, include: The acquisition module is used to acquire resource transfer channel data for the current resource transfer channel over multiple consecutive historical time ranges; The dynamic decision-making module is used to analyze the resource transfer channel data within the multiple historical time ranges through a pre-trained dynamic circuit breaker decision model to obtain the circuit breaker decision results and the predicted recovery strategy. The predicted recovery strategy is used to specify the recovery timing and traffic recovery rate of the current resource transfer channel. The recovery timing refers to the timing of restoring the current resource transfer channel after a preset time after the current resource transfer channel is circuit-broken. The traffic recovery rate refers to the rate at which traffic increases during the process of restoring the current resource transfer channel; The channel control module is configured to, when the circuit breaker decision result indicates that the current resource transfer channel is to be circuit breaker, to reduce the flow of the current resource transfer channel to a first flow rate, and to restore the current resource transfer channel at the time indicated by the recovery timing, according to the flow recovery rate.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 10.

13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

Images