Service calling link exception processing method, system and equipment and medium

Abstract

The invention discloses a service calling link exception handling method, system and device and a medium. The method comprises the steps of obtaining a target service calling link and determining current state vector information of the target service calling link; according to the current state vector information, in combination with a pre-trained exception handling network model, determining exception handling guidance information of the target service calling link; and determining a processing measure for the target service calling link according to the exception processing guidance information, and processing the target service calling link according to the processing measure. According to the method, the current state vector information is analyzed in combination with the exception handling network model, so that the exception handling guidance information is determined, and the processing measures for the target service calling link are determined according to the exception handling guidance information. The reliability and efficiency of identifying and processing potential abnormal conditions existing in the service calling link according to specific conditions are improved, the interruption time of the service is shortened, and therefore service continuity and good user experience are guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, system, device and medium for handling service call link exceptions. Background Art

[0002] With the rapid development of cloud computing and microservice architecture, the service call chain has become increasingly complex. This complexity has led to increased dependencies between services. The call relationship between services is usually dynamic, and the number of service instances will be adjusted in real time according to the load conditions, making it more difficult to monitor and troubleshoot service call chain anomalies.

[0003] At present, existing monitoring tools usually rely on static thresholds or rules, which are difficult to adapt to complex and changing environments. They are often unable to capture and analyze performance anomalies that may occur in the service call link in real time, resulting in delayed discovery of performance problems. In addition, existing anomaly detection methods mostly rely on rule-based mechanisms, which cannot flexibly adjust anomaly detection strategies according to specific circumstances and are difficult to deal with complex anomaly patterns, which in turn affects user experience and business continuity. Summary of the invention

[0004] The present invention provides a service call link exception processing method, system, device and medium to solve the problem that in the service call link exception processing process, the exception detection strategy cannot be flexibly adjusted according to the specific situation, resulting in inaccurate exception detection, and then leading to inaccurate response to the exception, affecting user experience and business continuity.

[0005] In a first aspect, an embodiment of the present invention provides a method for handling service call link exceptions, the method comprising:

[0006] Obtaining a target service call link and determining current state vector information of the target service call link;

[0007] Determine target exception handling guidance information of the target service call link according to the current state vector information and in combination with a pre-trained exception handling network model;

[0008] According to the target exception processing guidance information, a processing measure for the target service call link is determined, and the target service call link is processed according to the processing measure.

[0009] In a second aspect, an embodiment of the present invention provides a service call link exception processing system, the system comprising:

[0010] An information acquisition module, used to acquire a target service call link and determine current state vector information of the target service call link;

[0011] A guidance information determination module, configured to determine target exception handling guidance information of the target service call link according to the current state vector information and in combination with a pre-trained exception handling network model;

[0012] The exception handling module is used to determine a handling measure for a target service call link according to the target exception handling guidance information, and to handle the target service call link according to the handling measure.

[0013] In a third aspect, an embodiment of the present invention provides an electronic device, the electronic device comprising:

[0014] at least one processor;

[0015] and a memory communicatively coupled to the at least one processor;

[0016] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the service call link exception handling method described in any embodiment of the present invention.

[0017] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the service call link exception handling method described in any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention obtains the target service call link and determines the current state vector information of the target service call link; determines the exception handling guidance information of the target service call link based on the current state vector information and in combination with a pre-trained exception handling network model; determines the processing measures for the target service call link based on the exception handling guidance information, and processes the target service call link according to the processing measures. By analyzing the current state vector information in combination with the exception handling network model, the exception handling guidance information is determined, and the processing measures for the target service call link are determined based on the exception handling guidance information, thereby improving the reliability and efficiency of identifying and processing potential abnormal situations in the service call link according to specific circumstances, reducing the service interruption time, and thus ensuring business continuity and a good user experience.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flow chart of a method for handling service call link exceptions provided by an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of the structure of a service call link exception processing system provided by an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of the structure of an electronic device that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that existing monitoring tools usually rely on static thresholds or rules, and are often unable to capture and analyze performance anomalies that may occur in the service call link in real time, resulting in delayed discovery of performance problems. In addition, existing anomaly detection methods mostly rely on rule-based mechanisms and are unable to flexibly adjust anomaly detection strategies according to specific circumstances, resulting in inaccurate anomaly detection, which in turn leads to inaccurate responses to anomalies, affecting user experience and business continuity.

[0027] Based on this, an embodiment of the present invention provides a method for handling service call link exceptions. Figure 1 A flowchart of a service call link exception handling method provided in an embodiment of the present invention. The embodiment of the present invention is applicable to scenarios in which exception detection and processing of service call links are performed. The method can be executed by a service call link exception handling system, which can be implemented in the form of software and / or hardware, and optionally, by an electronic device, which is preferably a mobile terminal, a desktop computer, a laptop computer, a server, etc.

[0028] like Figure 1 As shown, the service call link exception processing method provided by the embodiment of the present invention may specifically include:

[0029] S101: Acquire a target service call link and determine current state vector information of the target service call link.

[0030] Among them, the target service call link can be understood as a service call link with anomaly detection requirements. The current state vector information can be understood as the vector information composed of the performance indicators and user feedback of the target service call link when processing requests in the current round after the target service call link is processed for exception handling, which is used to monitor the processing of the target service call link and evaluate the effect of handling the exceptions generated by the target service call link.

[0031] In this embodiment, the target service call link can be obtained based on information such as a trace identifier (trace id) and a span identifier (span id), and the current state vector information of the target service call link can be determined by collecting the performance indicator information of the target service call link in the process of processing the request.

[0032] Exemplarily, the current state vector information may include response time, error rate, request frequency, system load, saturation, user satisfaction, etc.

[0033] S102: Determine target exception handling guidance information of a target service call link based on current state vector information and in combination with a pre-trained exception handling network model.

[0034] The exception handling network model can be understood as a model for detecting abnormal situations in the target service call link and providing targeted processing instructions. The target exception handling guidance information can be understood as processing suggestions given for abnormal situations in the detected target service call link.

[0035] In this embodiment, the current state vector information of the target service call link is input into a pre-trained exception handling network model to determine whether the target service call link is abnormal, possible causes of the abnormality, and output corresponding target exception handling guidance information.

[0036] Exemplarily, when the target service call link status is judged to be normal or does not need adjustment according to the current state vector of the target service call link, the target exception handling guidance information is no operation; when the exception handling network model judges that the target service call link status is abnormal according to the current state vector of the target service call link, the determined target exception handling guidance information includes expanding resource allocation (such as increasing memory allocation or starting a new service instance, etc.), reducing resource allocation (such as reducing memory allocation, shutting down redundant service instances, etc.), restarting the service or calling a backup service, etc.

[0037] S103: Determine a processing measure for the target service call link according to the target exception processing guidance information, and process the target service call link according to the processing measure.

[0038] The processing measures may be understood as specific operations to be taken according to the target exception processing guidance information, which are used to improve or resolve the abnormal situation that occurs.

[0039] In this embodiment, according to the target exception handling guidance information, the processing measures for the target service call link are determined, wherein the processing measures can be input through the client or automatically triggered by the target exception handling guidance information. Then, the target service call link is processed accordingly according to the determined processing measures to improve or resolve the current abnormal situation, and then the current state vector information of the target service call link is re-determined, and S102-S103 is repeatedly executed until the monitoring cycle ends or there is no abnormal monitoring requirement for the target service call link.

[0040] A method for handling exceptions in a service call link provided by an embodiment of the present invention obtains a target service call link and determines the current state vector information of the target service call link; determines target exception handling guidance information of the target service call link based on the current state vector information and in combination with a pre-trained exception handling network model; determines handling measures for the target service call link based on the target exception handling guidance information, and handles the target service call link according to the handling measures. By analyzing the current state vector information in combination with the exception handling network model, the exception handling guidance information is determined, and the handling measures for the target service call link are determined based on the exception handling guidance information, thereby improving the reliability and efficiency of identifying and handling potential abnormal situations in the service call link according to specific circumstances, reducing the service interruption time, and thus ensuring business continuity and a good user experience.

[0041] As a first optional embodiment of this embodiment, based on the above embodiment, the step of determining the current state vector information of the target service call link may be concretized into the following steps:

[0042] a1) Obtaining performance indicator data collected by a data collection tool relative to the target service call link.

[0043] Among them, performance indicator data can be understood as various indicators that reflect the health status of the target service call link during operation, such as response time, throughput, error rate, service time, and customer satisfaction, etc., which are used to evaluate the performance of the target service call link to monitor abnormal conditions in the operation of the target service call link.

[0044] In this embodiment, the target service call link is determined, and performance indicator data collected relative to the target service call link is obtained from the performance indicator data of the service call link collected and stored using data collection tools, such as application performance management tools, service logs, and operating system monitoring tools.

[0045] b1) Preprocessing the performance indicator data, and determining the current state vector information of the target service call link in the constructed mathematical model according to the preprocessing result.

[0046] In this embodiment, the performance indicator data is preprocessed by cleaning, denoising and normalization, and mathematical modeling is performed on the exception handling problem of the target service call link. The indicator items required in the preprocessed performance indicators are determined by the constructed mathematical model, and features are extracted from the indicator items. The vector composed of the feature extraction results of each indicator is determined as the current state vector information of the target service call link.

[0047] The above technical scheme of this embodiment improves the accuracy and effectiveness of subsequent analysis of abnormal situations occurring in the target service call link and the corresponding target exception handling guidance information provided by preprocessing the acquired performance indicator data and determining the current state vector information of the target service call link through the constructed mathematical model.

[0048] As a second optional embodiment of this embodiment, the target exception handling guidance information of the target service call link can be determined based on the current state vector information and combined with a pre-trained exception handling network model, which is specifically optimized into the following steps:

[0049] a2) using the current state vector information as input of the exception handling network model to determine each expected benefit value corresponding to each exception handling guidance information under the current state vector information.

[0050] The expected benefit value can be understood as the value obtained by predicting the expected effect that different exception handling guidance information may bring under the current state vector information. The exception handling guidance information can be understood as the processing suggestions that can be taken for the abnormal situation. It should be noted that the exception handling guidance information can be determined by mathematical modeling of the exception handling problem of the target service call link.

[0051] In this embodiment, the current state vector information is input into the exception handling network model, and the exception handling network model determines the expected profit values ​​corresponding to different exception handling guidance information according to the current state vector information.

[0052] b2) determining target exception handling guidance information of the target service call link output by the exception handling network model according to the expected benefit values.

[0053] In this embodiment, the maximum expected benefit value is determined by comparing the expected benefit values, and the exception handling guidance information corresponding to the maximum expected benefit value is determined as the target exception handling guidance information of the target service call link output by the exception handling network model.

[0054] The above technical scheme of this embodiment uses the exception handling network model to determine the expected benefit values ​​corresponding to each exception handling guidance information under the current state vector information, and determines the exception handling guidance information corresponding to the maximum expected benefit value as the target exception handling guidance information output by the exception handling network model, thereby improving the accuracy of the target exception handling guidance information given for the abnormal situation occurring in the target service call link, and improving the ability to effectively respond to various abnormal situations, and reducing the dependence on static thresholds.

[0055] As a third optional embodiment of this embodiment, the step of determining a processing measure for a target service call link according to the target exception processing guidance information, and processing the target service call link according to the processing measure may be specifically implemented as the following steps:

[0056] a3) Obtaining predetermined manual processing information.

[0057] The manual processing information may be understood as the exception processing guidance information corresponding to the situations that require manual confirmation, and is used to clarify the situations that require manual intervention.

[0058] For example, restarting a service may affect the ongoing business process and user experience, and requires manual confirmation on whether to execute it. Therefore, it is necessary to determine in advance that the manual processing information includes restarting the service.

[0059] In this embodiment, manual processing information that is determined and stored in advance according to needs is obtained.

[0060] b3) If the target exception processing guidance information is the same as the manual processing information, a first processing measure for the target service call link is determined, and the target service call link is processed accordingly according to the first processing measure.

[0061] The first processing measure may be understood as a processing measure instruction input by the client.

[0062] In this embodiment, if the target exception handling guidance information is the same as the manual processing information, it is determined that the target exception handling guidance information given for the abnormal situation occurring in the current target service call link needs to be manually determined, and then the processing measure instruction input by the client is obtained, and the processing measure instruction is determined as the first processing measure for the target service call link, and the target service call link is processed according to the first processing measure.

[0063] c3) If the target exception processing guidance information is different from the manual processing information, determining a second processing measure for the target service call link, and processing the target service call link according to the second processing measure.

[0064] The second processing measure may be understood as a measure instruction triggered by the target exception processing guidance information and automatically processed according to the target exception processing guidance information.

[0065] In this embodiment, if the target exception handling guidance information is different from the manual processing information, it is determined that the target exception handling guidance information given for the abnormal situation occurring in the current target service call link does not need to be manually determined. At this time, the second processing measure for the target service call link is determined according to the target exception handling guidance information, that is, the processing measure corresponding to the exception handling guidance information, and the target service call link is processed accordingly according to the second processing measure.

[0066] The above technical solution of this embodiment determines the manual processing information in advance and judges whether the target exception processing guidance information is manual processing information, and determines the corresponding processing measures to ensure that when an exception occurs in the service call link, it can respond safely, quickly and effectively to reduce the impact on the business and user experience, and reduce the dependence on manual intervention and reduce operation and maintenance costs, especially in the management of large-scale service call links.

[0067] As a fourth optional embodiment of this embodiment, based on the above embodiment, it also includes:

[0068] a4) After processing the target service call link according to the processing measure, obtaining next state vector information and determining a reward function value of a given reward function.

[0069] Among them, the next state vector information can be understood as the round after the processing measures (such as restarting the service, shutting down redundant service instances, etc.) are executed, and the performance indicators and user feedback of the target service call link when processing the request are formed. The reward function can be understood as a function used to guide the behavior of the exception handling network model. The reward function value can be understood as the value of evaluating the processing measures, which is used to help the exception handling network model gradually learn the optimal processing measures.

[0070] In this embodiment, the next state vector information of the target service call link after the target service call link is processed according to the processing measure is obtained, and the reward function value of the given reward function is determined.

[0071] It can be understood that if the processing measures taken effectively improve the performance of the target service call link, such as reducing the response time and lowering the error rate, a positive reward is given, that is, the reward function value is positive; if the processing measures taken cause the performance of the target service call link to deteriorate, such as increasing the response time and increasing the error rate, a negative reward is given, that is, the reward function value is negative.

[0072] For example, the reward function can be expressed as:

[0073] R=α 1 *(T t-1 -T t )+α 2 *(E t-1 -E t )+α 2 *(η t -η t-1 );

[0074] Among them, R is the reward function value, α 1 , α 2 and α 3 The weight coefficients are used to measure the response time, error rate, and resource utilization efficiency of the target service call link. t-1 and T t are the response time of the previous round and the response time of the current round of the target service call link, respectively. t-1 and E t are the error rate of the previous round and the error rate of the current round of the target service call link, respectively, t-1 and η t They are the resource usage rate of the previous round and the resource usage rate of the current round of the target service call link respectively.

[0075] b4) The determined current state vector information, the target exception handling guidance information, the next state vector information and the reward function value are stored in a database as sample data for training the exception handling network model.

[0076] In this embodiment, the current state vector information, target exception handling guidance information, next state vector information and reward function value determined in each round of the target service call link are stored in a database or other storage system, and a training cycle is set to use the stored data as sample data for subsequent model training of the exception handling network model.

[0077] The above technical solution of this embodiment obtains the next state vector information and the reward function value after processing the target service call link according to the processing measures, and stores the current state vector information, the target exception processing guidance information, and the next state vector information and the reward function value in the database as sample data for training the exception processing network model, which helps the exception processing network model to efficiently improve the adaptive ability, accuracy and efficiency of identifying and responding to potential exceptions according to specific circumstances, thereby further enhancing the user experience.

[0078] As a fifth optional embodiment of this embodiment, based on the above embodiment, it also includes:

[0079] a5) After determining the target exception handling guidance information, generate and send an alarm message.

[0080] In this embodiment, after the target exception handling guidance information is determined, detailed alarm information is generated, wherein the alarm information includes the abnormal nature of the exception generated by the target service call link, the affected services, and the target exception handling guidance information, and then the generated alarm information is sent via email, text message, etc.

[0081] The above technical solution of this embodiment, after determining the target exception handling guidance information, generates and sends alarm information, reduces the burden of manual monitoring, and effectively improves the efficiency and accuracy of exception handling by providing the nature of the exception, the affected services and the target exception handling guidance information.

[0082] As a sixth optional embodiment of this embodiment, the training of the exception handling network model may be concretized into the following steps:

[0083] a6) Determine sample state vector information of at least one service call link based on the stored historical data information.

[0084] The historical data can be understood as the current state vector information, target exception handling guidance information, next state vector information, and reward function value of each service call link in different rounds. The sample state vector information can be understood as the current state vector information of the sample used to train the exception handling network model.

[0085] In this embodiment, sample state vector information of at least one service call link is determined based on the stored historical data information of each service call link.

[0086] b6) Inputting the sample state vector information into the exception handling network model to obtain the sample expected return value and sample exception handling guidance information.

[0087] The sample expected return value can be understood as the maximum expected return value determined according to the sample state vector information, and the sample abnormality handling guidance information can be understood as the abnormality handling guidance information corresponding to the sample expected return value.

[0088] In this embodiment, the sample state vector information is input into the exception handling network model to obtain the sample expected return value determined by the exception handling network model and the corresponding output sample exception handling guidance information.

[0089] c6) Processing the service call link according to the sample processing measures corresponding to the sample exception processing guidance information, and obtaining the next sample state vector information.

[0090] The sample processing measures can be understood as the corresponding processing measures given by the sample exception processing guidance information. The next sample state vector information can be understood as the state vector information obtained in the round after the service call link is processed accordingly.

[0091] In this embodiment, the corresponding sample processing measures are determined according to the sample exception processing guidance information, and the service call link is processed according to the sample processing measures, and then the next sample state vector information of the processed service call link is obtained.

[0092] d6) determining a sample reward function value of a given reward function based on the sample state vector information and the next sample state vector information.

[0093] Among them, the sample reward function value can be considered as the value obtained according to the given reward function during the training of the exception handling network model, which is used to help the model adjust the strategy of identifying and responding to exceptions, so as to gradually learn the optimal sample processing measures in different scenarios.

[0094] e6) Inputting the next sample state vector information into the target network model to obtain the target expected return value, and obtaining the updated sample expected return value based on the sample expected return value, the sample reward function value and the target expected return value in combination with a given target optimization function.

[0095] Among them, the target network model can be understood as a model used to provide a target value for the exception handling network model to guide the exception handling network to converge to the optimal strategy. The target expected return value can be understood as the maximum value of the sample expected return value obtained in the case of the next sample state vector information determined by the target network model. The target optimization function can be considered as a function used to adjust the sample expected return value output by the exception handling network model so that the sample expected return value is closer to the target expected return value. The updated sample expected return value can be understood as the new sample expected return value obtained according to the target optimization function.

[0096] It should be noted that the model parameters of the target network model are determined by obtaining the network parameters corresponding to the maximum sample expected return value historically obtained in the exception handling network model at set time intervals, so as to ensure that the output accuracy of the target network model increases steadily.

[0097] In this embodiment, the next sample state vector information is input into the target network model to obtain the target expected return value. Through the set target optimization function, the sample expected return value output by the exception processing network model is updated according to the sample expected return value, the sample reward function value and the target expected return value to obtain the updated sample expected return value.

[0098] Exemplarily, the objective optimization function can be expressed as:

[0099] Q'(s,a)=Q(s,a)+β(r+γmax a' Q target (s',a')-Q(s,a));

[0100] Among them, s is the sample state vector information, a is the sample anomaly handling guidance information, Q(s,a) is the sample expected return value, Q'(s,a) is the updated sample expected return value, β is the learning rate, r is the sample reward function value, γ is the discount factor, s' is the next sample state vector information, a' is the sample anomaly handling guidance information obtained by the target network model according to the next sample state vector information, Q target (s',a') is the target expected return value.

[0101] f6) updating the network parameters of the exception handling network model according to the sample expected return value and the updated sample expected return value in combination with a given target loss function.

[0102] In this embodiment, according to the sample expected return value and the updated sample expected return value, the network parameters of the exception processing network model are updated through a given target loss function, such as a mean square error function, to obtain updated network parameters, and the next sample state vector information is used as the new sample state vector information, and steps b6)-f6) are re-executed until the exception processing network model reaches convergence conditions.

[0103] Continuing with the above description, during the training process of the exception handling network model, the sample state vector information, the sample exception handling guidance information, the sample expected return value and the next sample state vector information are stored, so that in subsequent training, the time correlation of the training samples can be broken by obtaining part of the above stored data.

[0104] The above technical solution of this embodiment reduces the transmission of estimation errors and the impact of time-related training samples on the exception handling network model by combining the target network model, a given target optimization function and a given target loss function based on the stored historical data information and the information stored during the sample training process, thereby improving the training stability, efficiency and accuracy of the exception handling network model.

[0105] Figure 2 The structure diagram of a service call link exception handling system provided by an embodiment of the present invention is as follows. Figure 2 As shown, the system includes: an information acquisition module 21, a guidance information determination module 22 and an exception handling module 23, wherein:

[0106] An information acquisition module 21 is used to acquire a target service call link and determine current state vector information of the target service call link;

[0107] A guidance information determination module 22, configured to determine target exception handling guidance information of the target service call link according to the current state vector information and in combination with a pre-trained exception handling network model;

[0108] The exception handling module 23 is used to determine a handling measure for the target service call link according to the target exception handling guidance information, and handle the target service call link according to the handling measure.

[0109] A service call link exception processing system provided by an embodiment of the present invention obtains a target service call link and determines the current state vector information of the target service call link; determines the target exception processing guidance information of the target service call link based on the current state vector information and in combination with a pre-trained exception processing network model; determines the processing measures for the target service call link based on the target exception processing guidance information, and processes the target service call link according to the processing measures. By analyzing the current state vector information in combination with the exception processing network model, the exception processing guidance information is determined, and the processing measures for the target service call link are determined based on the exception processing guidance information, thereby improving the reliability and efficiency of identifying and processing potential abnormal situations existing in the service call link according to specific circumstances, reducing the service interruption time, and thus ensuring business continuity and a good user experience.

[0110] Furthermore, the information acquisition module 21 can be specifically used for:

[0111] Obtaining performance indicator data collected by a data collection tool relative to the target service call link;

[0112] The performance indicator data is preprocessed, and the current state vector information of the target service call link is determined in the constructed mathematical model according to the preprocessing result.

[0113] Furthermore, the guidance information determination module 22 may be specifically used for:

[0114] Using the current state vector information as input of the exception handling network model, determining each expected benefit value corresponding to each exception handling guidance information under the current state vector information;

[0115] According to the expected benefit values, target exception handling guidance information of the target service call link output by the exception handling network model is determined.

[0116] Furthermore, the exception handling module 23 can be specifically used for:

[0117] Obtain preset manual processing information;

[0118] If the target exception processing guidance information is the same as the manual processing information, determining a first processing measure for the target service call link, and processing the target service call link according to the first processing measure;

[0119] If the target exception processing guidance information is different from the manual processing information, a second processing measure for the target service call link is determined, and the target service call link is processed according to the second processing measure.

[0120] Furthermore, the service call link exception handling system also includes:

[0121] After processing the target service call link according to the processing measure, obtaining next state vector information and determining a reward function value of a given reward function;

[0122] The determined current state vector information, the target exception handling guidance information, the next state vector information and the reward function value are stored in a database to serve as sample data for training the exception handling network model.

[0123] After determining the target exception handling guidance information, an alarm message is generated and sent.

[0124] Furthermore, the service call link exception handling system also includes a network training module, which can be specifically used for:

[0125] Determine sample state vector information of at least one service call link according to the stored historical data information;

[0126] Inputting the sample state vector information into the exception handling network model to obtain the sample expected return value and sample exception handling guidance information;

[0127] Process the service call link according to the sample processing measure corresponding to the sample exception processing guidance information, and obtain next sample state vector information;

[0128] Determining a sample reward function value of a given reward function according to the sample state vector information and the next sample state vector information;

[0129] Input the next sample state vector information into the target network model to obtain the target expected return value, and obtain the updated sample expected return value based on the sample expected return value, the sample reward function value and the target expected return value in combination with a given target optimization function;

[0130] According to the sample expected return value and the updated sample expected return value, combined with a given target loss function, the network parameters of the exception processing network model are updated.

[0131] The service call link exception processing system provided in the embodiment of the present invention can execute the service call link exception processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0132] Figure 3A schematic diagram of the structure of an electronic device 30 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0133] like Figure 3 As shown, the electronic device 30 includes at least one processor 31, and a memory connected to the at least one processor 31, such as a read-only memory (ROM) 32, a random access memory (RAM) 33, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 31 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 32 or the computer program loaded from the storage unit 38 to the random access memory (RAM) 33. In the RAM 33, various programs and data required for the operation of the electronic device 30 can also be stored. The processor 31, the ROM 32, and the RAM 33 are connected to each other through a bus 34. An input / output (I / O) interface 35 is also connected to the bus 34.

[0134] A number of components in the electronic device 30 are connected to the I / O interface 35, including: an input unit 36, such as a keyboard, a mouse, etc.; an output unit 37, such as various types of displays, speakers, etc.; a storage unit 38, such as a disk, an optical disk, etc.; and a communication unit 39, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 39 allows the electronic device 30 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0135] The processor 31 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 31 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 31 executes the various methods and processes described above, such as a service call link exception processing method.

[0136] In some embodiments, the service call link exception handling method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 38. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 30 via the ROM 32 and / or the communication unit 39. When the computer program is loaded into the RAM 33 and executed by the processor 31, one or more steps of the service call link exception handling method described above may be performed. Alternatively, in other embodiments, the processor 31 may be configured to execute the service call link exception handling method in any other appropriate manner (e.g., by means of firmware).

[0137] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0138] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0139] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0140] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0141] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0142] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0143] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0144] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for handling service call link exceptions, characterized in that: include: Obtaining a target service call link and determining current state vector information of the target service call link; Determine target exception handling guidance information of the target service call link according to the current state vector information and in combination with a pre-trained exception handling network model; According to the target exception processing guidance information, a processing measure for the target service call link is determined, and the target service call link is processed according to the processing measure.

2. The method according to claim 1, characterized in that: The determining of the current state vector information of the target service call link includes: Obtaining performance indicator data collected by a data collection tool relative to the target service call link; The performance indicator data is preprocessed, and the current state vector information of the target service call link is determined in the constructed mathematical model according to the preprocessing result.

3. The method according to claim 1, characterized in that: The determining, based on the current state vector information and in combination with a pre-trained exception handling network model, target exception handling guidance information of the target service call link includes: Using the current state vector information as input of the exception handling network model, determining each expected benefit value corresponding to each exception handling guidance information under the current state vector information; According to the expected benefit values, target exception handling guidance information of the target service call link output by the exception handling network model is determined.

4. The method according to claim 1, characterized in that: The determining, according to the target exception handling guidance information, a handling measure for the target service call link, and processing the target service call link according to the handling measure, includes: Obtain preset manual processing information; If the target exception processing guidance information is the same as the manual processing information, determining a first processing measure for the target service call link, and processing the target service call link according to the first processing measure; If the target exception processing guidance information is different from the manual processing information, a second processing measure for the target service call link is determined, and the target service call link is processed according to the second processing measure.

5. The method according to claim 1, characterized in that Also includes: After processing the target service call link according to the processing measure, obtaining next state vector information and determining a reward function value of a given reward function; The determined current state vector information, the target exception handling guidance information, the next state vector information and the reward function value are stored in a database to serve as sample data for training the exception handling network model.

6. The method according to claim 1, characterized in that Also includes: After determining the target exception handling guidance information, an alarm message is generated and sent.

7. The method according to claim 1, characterized in that The training steps of the exception handling network model include: Determine sample state vector information of at least one service call link according to the stored historical data information; Inputting the sample state vector information into the exception handling network model to obtain the sample expected return value and sample exception handling guidance information; Process the service call link according to the sample processing measure corresponding to the sample exception processing guidance information, and obtain next sample state vector information; Determining a sample reward function value of a given reward function according to the sample state vector information and the next sample state vector information; Input the next sample state vector information into the target network model to obtain the target expected return value, and obtain the updated sample expected return value based on the sample expected return value, the sample reward function value and the target expected return value in combination with a given target optimization function; According to the sample expected return value and the updated sample expected return value, combined with a given target loss function, the network parameters of the exception processing network model are updated.

8. A service call link exception handling system, characterized in that: include: An information acquisition module, used to acquire a target service call link and determine current state vector information of the target service call link; A guidance information determination module, configured to determine target exception handling guidance information of the target service call link according to the current state vector information and in combination with a pre-trained exception handling network model; The exception handling module is used to determine a handling measure for a target service call link according to the target exception handling guidance information, and to handle the target service call link according to the handling measure.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the service call link exception handling method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the service call link exception handling method described in any one of claims 1 to 7 when executed.

Images