Network performance monitoring method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of Internet, and discloses a network performance monitoring method and device, computer equipment and a storage medium, and the method comprises the steps: determining change information of a current monitoring page and a current monitoring element according to a user operation instruction, and respectively determining waiting duration information and operation duration information; and generating network performance prediction data through a machine learning algorithm, the waiting duration information and the operation duration information. By means of the mode, the current monitoring page and the current monitoring element are determined, accurate monitoring of the whole process of user operation is achieved, waiting time and potential performance problems caused by response delay are accurately recognized, possible bottlenecks or abnormities are predicted, and user experience is improved. Related digital financial institutions or digital medical institutions can carry out resource allocation and optimization in advance according to the prediction result, so that the network performance monitoring capability in the financial science and technology business field and the digital medical field is improved.

Description

Technical Field

[0001] The present application relates to the field of Internet technology and can be applied to the field of financial technology business, and in particular to a network performance monitoring method, device, computer equipment and storage medium. Background Art

[0002] With the rapid development of internet technology, digital finance has become an integral part of modern financial services. Digital financial platforms leverage technologies such as the internet, mobile applications, and cloud computing to provide users with convenient financial services such as online payment, mobile banking, digital currency trading, and robo-advisory. The efficiency and stability of these services are directly related to user experience, the reputation of financial institutions, and the stability of the financial markets.

[0003] In digital healthcare, hospital information systems such as electronic medical records (EMRs), picture acquisition systems (PACS), and telemedicine service platforms also have extremely stringent performance requirements. For example, when reviewing patient records, medical staff must quickly load large amounts of text, images, and other diverse data. During telemedicine consultations, the stable transmission of high-definition video streams and the real-time interaction of relevant examination data all rely on the system's efficient performance.

[0004] Currently, existing performance monitoring platforms primarily implement their monitoring functions by collecting application performance data. Users can view system performance through visualization platforms (such as charts and dashboards). Furthermore, these platforms typically monitor performance indicators based on preset thresholds and rules, triggering alarms when performance exceeds these thresholds. This monitoring approach, to a certain extent, helps operations personnel promptly identify and address performance issues.

[0005] However, the complexity and diversity of digital financial services and digital healthcare services present new challenges. Financial transactions require high real-time performance, while healthcare prioritizes stability and data timeliness. Any performance bottleneck or delay can lead to transaction failures and a degraded user experience. Therefore, improving network performance monitoring capabilities in both the fintech and digital healthcare sectors has become a pressing technical challenge. Summary of the Invention

[0006] The present application provides a network performance monitoring method, apparatus, computer equipment, and storage medium to improve network performance monitoring capabilities in the financial technology business field and the digital healthcare field.

[0007] In a first aspect, the present application provides a network performance monitoring method, the method comprising:

[0008] In the case where a network request is triggered by a user operation instruction, determining a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction;

[0009] Monitor change information of the current monitoring page and the current monitoring element, and determine waiting time information and operation time information respectively according to the change information;

[0010] Network performance prediction data is generated by using a preset machine learning algorithm, the waiting time information, and the operation time information.

[0011] In a second aspect, the present application further provides a network performance monitoring device, the device comprising:

[0012] A monitoring target determination module is used to determine a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction when a network request is triggered by the user operation instruction;

[0013] a duration information determination module, configured to monitor change information of the current monitoring page and the current monitoring element, and determine waiting duration information and operation duration information respectively according to the change information;

[0014] The network performance prediction data generation module is used to generate network performance prediction data through a preset machine learning algorithm, the waiting time information and the operation time information.

[0015] In a third aspect, the present application also provides a computer device, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the network performance monitoring method as described above when executing the computer program.

[0016] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the network performance monitoring method as described above.

[0017] The present application discloses a network performance monitoring method, apparatus, computer equipment and storage medium, wherein the network performance monitoring method includes determining the current monitoring page and the current monitoring element in the current monitoring page according to the user operation instruction when a network request is triggered by a user operation instruction; monitoring the change information of the current monitoring page and the current monitoring element, and determining the waiting time information and the operation time information respectively according to the change information; and generating network performance prediction data by presetting a machine learning algorithm, the waiting time information and the operation time information. In the above manner, the present application realizes accurate monitoring of the entire user operation process by identifying the network request triggered by the user operation instruction and determining the current monitoring page and the current monitoring element, accurately identifying the waiting time caused by the user's response delay during the operation, identifying potential performance problems in advance, and predicting bottlenecks or anomalies that may occur in the future. Relevant digital financial institutions can allocate and optimize resources in advance based on the prediction results, thereby improving the network performance monitoring capabilities in the digital financial and digital medical fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 is a schematic flow chart of a network performance monitoring method provided by the first embodiment of the present application;

[0020] Figure 2 is a schematic flow chart of a network performance monitoring method provided by the second embodiment of the present application;

[0021] Figure 3 A schematic block diagram of a network performance monitoring device provided in an embodiment of the present application;

[0022] Figure 4 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0025] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0026] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0027] The embodiments of the present application provide a network performance monitoring method, apparatus, computer equipment, and storage medium. Specifically, the network performance monitoring method can be applied to a server. By identifying network requests triggered by user operation instructions and determining the current monitoring page and the current monitoring element, accurate monitoring of the entire user operation process can be achieved. The waiting time caused by response delays during the user's operation can be accurately identified, potential performance issues can be identified in advance, and bottlenecks or anomalies that may occur in the future can be predicted. Relevant digital financial institutions can allocate and optimize resources in advance based on the prediction results, thereby improving the network performance monitoring capabilities in the digital financial and digital medical fields. Specifically, the server can be an independent server or a server cluster.

[0028] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0029] See also Figure 1 , Figure 1 This is a schematic flow chart of a network performance monitoring method provided by the first embodiment of the present application. This network performance monitoring method can be applied to a server to accurately monitor the entire user operation process by identifying network requests triggered by user operation instructions and determining the current monitoring page and current monitoring element. It accurately identifies the waiting time caused by response delays during the user operation process, identifies potential performance issues in advance, and predicts bottlenecks or anomalies that may occur in the future. Relevant digital financial institutions can allocate and optimize resources in advance based on the prediction results, thereby improving the network performance monitoring capabilities in the digital financial and digital medical fields.

[0030] like Figure 1As shown, the network performance monitoring method specifically includes steps S10 to S30.

[0031] Step S10: When a network request is triggered by a user operation instruction, determining a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction;

[0032] In one embodiment, in a front-end page (such as a browser or mobile application), a performance monitoring SDK (Software Development Kit) is used to capture user operation instructions based on an event monitoring mechanism, such as clicking a button, submitting a form, or jumping to a page.

[0033] According to the user operation instructions, the current monitoring page (such as the current URL (Uniform Resource Locator) or page ID) and the page elements that trigger the operation can be obtained through DOM (read document) operations or the state management of the front-end framework. The page elements can be button IDs, form fields, etc. When the user operation instruction triggers a network request, the initiation time of the request and the detailed information of the request are recorded. The detailed information can be the interface URL, request method, request parameters, etc. The status changes of the monitored page include the time when the page is loaded, the changes of the page elements (such as displaying the loading status, showing / hiding the global mask layer), or the time when the page rendering is completed. The status changes of the monitored page elements include whether the button status changes, whether the loading animation or progress bar is displayed, the response time of the element, etc.

[0034] Step S20: monitoring change information of the current monitoring page and the current monitoring element, and determining waiting time information and operation time information respectively according to the change information;

[0035] Specifically, the waiting time information is the time a user must wait before performing the next operation when performing a certain operation, and the operation time information is the total time a user spends performing a certain operation (such as request and page rendering, etc.).

[0036] Identify whether the currently monitored page hinders user operations, such as displaying loading or a global mask layer, and calculate the time difference from display to hiding, which is recorded as waiting time information; combine the page loading time, element response time and network request time to calculate the operation duration information.

[0037] Step S30: Generate network performance prediction data through a preset machine learning algorithm, the waiting time information, and the operation time information.

[0038] Specifically, network performance data is collected, including wait time and operation time information. The wait and operation time data is normalized to facilitate model training. These wait and operation time information are used as input features and fed into a pre-trained machine learning algorithm.

[0039] Through pre-configured machine learning, we can determine the time a user request spends in the network queue and the actual execution time of the task (e.g., data transmission time, server processing time), identifying network performance indicators (e.g., network latency, network utilization, or packet loss rate). We can perform single-step or multi-step predictions on wait and operation duration information, then use pre-configured machine learning algorithms to detect outliers or unusual patterns, and eliminate outliers such as fluctuations caused by poor user network conditions.

[0040] By analyzing historical performance data, we can identify trends and patterns in performance indicators and predict future data changes. We can also analyze the impact of correlations, the series of methods called by a user in a single operation, discover the mutual impact among them, and find the root causes that may lead to performance problems.

[0041] This embodiment discloses a network performance monitoring method, which includes, when a network request is triggered by a user operation instruction, determining the current monitoring page and the current monitoring element in the current monitoring page according to the user operation instruction; monitoring the change information of the current monitoring page and the current monitoring element, and determining the waiting time information and the operation time information respectively according to the change information; generating network performance prediction data by presetting a machine learning algorithm, the waiting time information and the operation time information. In the above manner, the present application realizes accurate monitoring of the entire user operation process by identifying the network request triggered by the user operation instruction and determining the current monitoring page and the current monitoring element, accurately identifying the waiting time caused by the user's response delay during the operation, identifying potential performance problems in advance, and predicting bottlenecks or anomalies that may occur in the future. Relevant digital financial institutions can allocate and optimize resources in advance based on the prediction results, thereby improving the network performance monitoring capabilities in the digital financial and digital medical fields.

[0042] See also Figure 2 , Figure 2This is a schematic flow chart of a network performance monitoring method provided by the second embodiment of the present application. This network performance monitoring method can be applied to the server, covering the interaction of the front-end page and also going deep into the call of the back-end interface, ensuring that the monitoring covers every link from the user initiating the request to the system response. Identifying the waiting time caused by the user's response delay during the operation process and the total time required to complete the entire operation, quickly locating the root cause of the performance problem, identifying potential abnormal patterns, and providing strong support for subsequent performance optimization, thereby improving the network performance monitoring capabilities in the digital financial and digital medical fields.

[0043] based on Figure 1 The embodiment shown, this embodiment Figure 2 As shown, step S30 includes steps S301 to S303.

[0044] Step S301: Acquire historical network performance information, and extract historical performance feature vectors from the historical network performance information;

[0045] Specifically, the historical network performance information obtained may include interface response time, user wait time, operation execution time, network latency, system load, page load time, or server response status code. Feature extraction is performed on the historical network performance data to generate a historical performance feature vector. The historical performance feature vector may include average response time, maximum response time, average wait time, average operation time, percentile of network latency, or average system load. The extracted features are organized into a feature matrix, where each row represents the performance status at a point in time and each column represents a feature variable.

[0046] Step S302: determining network abnormal fluctuation information through the preset machine learning algorithm, the waiting time information, and the operation time information, and constructing an abnormal fluctuation feature matrix of the network abnormal fluctuation information;

[0047] Specifically, we extract basic features related to network performance, such as average wait time, maximum operation duration, average interface response time, percentile of network latency, or average system load. We organize the extracted features into a feature matrix, where each row represents the performance status at a point in time and each column represents a feature variable.

[0048] Step S303: Generate the network performance prediction data according to the historical performance characteristic vector and the abnormal fluctuation characteristic matrix.

[0049] Specifically, the trained model is used to analyze monitoring data and identify abnormal fluctuations. Abnormal fluctuations can include sudden increases in interface response time, abnormal peaks in wait time, or sudden increases in network latency. Detected abnormal fluctuations are marked, and their timestamps, abnormal feature values, and anomaly types are recorded.

[0050] In one embodiment, this embodiment is illustrated in conjunction with the field of digital financial services. Online trading systems (such as stock trading, foreign exchange trading, and digital currency trading) need to process a large number of transaction requests in real time. Fluctuations in network performance may cause transaction delays, timeouts, or failures, thereby affecting user experience and transaction success rates. By monitoring network requests triggered by user operation instructions, waiting time and operation time information are obtained in real time, and machine learning algorithms are used to detect abnormal fluctuations and identify performance bottlenecks that may lead to transaction failures. Based on historical performance feature vectors and abnormal fluctuation feature matrices, future network performance is predicted and system resources are optimized in advance.

[0051] In another example, a financial institution relies on a data analytics platform for market analysis, risk assessment, and investment decision-making. Financial analytics platforms must rapidly process large amounts of data, and any performance bottlenecks can lead to decision delays. This example monitors network requests in the data processing flow, records wait times and operation durations, and uses machine learning algorithms to detect abnormal fluctuations in data processing. This allows for timely warnings and, based on historical data and abnormal fluctuation characteristics, predicts future performance and optimizes the data processing flow. By predicting and optimizing network performance, transaction failures due to system delays or errors are reduced, interface response times and system resource allocation are optimized, and transaction processing speed is increased.

[0052] This embodiment discloses a network performance monitoring method, which includes determining a current monitoring page and a current monitoring element in the current monitoring page according to a user operation instruction when a network request is triggered by the user operation instruction; monitoring change information of the current monitoring page and the current monitoring element, and determining wait time information and operation time information respectively according to the change information; obtaining historical network performance information and extracting historical performance feature vectors from the historical network performance information; determining network abnormal fluctuation information using the preset machine learning algorithm, the wait time information and the operation time information, and constructing an abnormal fluctuation feature matrix of the network abnormal fluctuation information; and generating network performance prediction data based on the historical performance feature vectors and the abnormal fluctuation feature matrix. Through the above-mentioned method, this application covers the interaction of the front-end page and also goes deep into the call of the back-end interface, ensuring that monitoring covers every link from user initiation of the request to system response. Identifying the waiting time caused by system response delay during the user operation and the total time required to complete the entire operation, quickly locating the root cause of performance problems, identifying potential abnormal patterns, providing strong support for subsequent performance optimization, and thus improving the network performance monitoring capabilities in the digital financial and digital medical fields.

[0053] based on Figure 2 In the embodiment shown, in this embodiment, step S303 includes:

[0054] Fusing the historical performance feature vector with the abnormal fluctuation feature matrix to generate a comprehensive feature matrix;

[0055] The comprehensive feature matrix is ​​recursively predicted by the preset machine learning algorithm to generate the network performance prediction data.

[0056] In one embodiment, performance features extracted from historical data, such as average response time, average wait time, and average operation time, and abnormal features extracted from an abnormal fluctuation feature matrix, such as abnormal response time, abnormal wait time, and abnormal operation time, are aligned with the timestamps of the abnormal fluctuation feature matrix to merge the features at each time point. If the timestamps are inconsistent, alignment can be achieved through interpolation or filling in missing values.

[0057] The fusion processing of the historical performance characteristic vector and the abnormal fluctuation characteristic matrix can be carried out through splicing, weighted fusion or feature conversion. Specifically, a more appropriate fusion method can be determined according to the actual situation of the historical performance characteristic vector and the abnormal fluctuation characteristic matrix.

[0058] Using the comprehensive feature matrix as input, train the preset machine learning model, divide the data into a training set and a test set (for example, 80% training set, 20% test set), train the model using the training set data, and verify the model performance using the test set. Use the features of the last time point in the test set as the initial input, input the features of the current time point into the model, generate the predicted value for the next time point, use the predicted value as the new feature input, and continue to predict the next time point. Repeat the above steps until a prediction sequence of the desired length is generated. Convert the recursive prediction results into interpretable performance indicators, such as predicted interface response time, predicted user waiting time, predicted operation execution time, etc.

[0059] based on Figure 1 In the illustrated embodiment, step S20 includes:

[0060] Determining whether a preset loading element exists on the current monitoring page according to the change information of the current monitoring element;

[0061] In a case where the preset loading element exists on the current monitoring page, the waiting time information is determined.

[0062] In one embodiment, the preset loading element can be a label displaying "Loading" or other identifier indicating that the page is loading. All key financial operations involving page loading are listed, including payment confirmation, transaction submission, risk assessment, and real-time data refresh. A unique loading element identifier is defined for each operation, such as payment confirmation, transaction submission, and report generation. A loading completion flag is set for each key loading element.

[0063] In one embodiment, when a user submits a transaction request in an online transaction system, a loading progress bar will be displayed on the page until the transaction result is returned. In this scenario, the loading progress bar is a pre-set loading element. If the currently monitored page contains pre-set loading elements, further confirmation is performed to determine whether these elements have been loaded. When the user triggers the operation, the loading start time is recorded. When the pre-set loading element is loaded, the loading completion time is recorded. Waiting time information is further calculated based on the loading start and completion times.

[0064] In one embodiment, this embodiment can also be applied to the field of digital medical services. The following describes this embodiment in conjunction with an application scenario in the field of digital medical services.

[0065] In the digital healthcare sector, for example, the performance of an online medical consultation app is crucial to user experience and the timeliness and reliability of medical services. Online medical consultation applications involve front-end mobile applications or web pages, back-end medical service systems such as doctor consultation systems, medical databases, payment systems, and network communications.

[0066] In actual applications, online medical consultation applications use SDKs to monitor user click operations in the application. For example, when a user clicks the "Initiate Medical Consultation" button, the "Upload Symptom Image" button, or the "View Historical Diagnosis Records" button, etc. When a click event occurs, the name of the page triggered by the click is recorded, such as the medical consultation homepage, the symptom upload page, the diagnosis record list page, and the specific element name (such as the "Initiate Medical Consultation" button ID);

[0067] When the user clicks the "Initiate Consultation" button, the online consultation application will send a network request to the medical backend service system to create a consultation record; when uploading symptom pictures, it will send a request to upload picture data; when viewing historical diagnosis records, it will send a request to obtain a list of diagnosis records, etc.

[0068] After the user initiates a consultation, the online consultation application will display a loading animation to prompt the user that the consultation information is being submitted. During this period, the user cannot perform other operations. The SDK monitors page changes, identifies the moment when the loading animation is displayed, and the moment when the loading animation is hidden. The time difference between the two is calculated, which is the user's operational time. For example, the time from displaying the loading animation to hiding the loading animation is 3 seconds, so the user's waiting time information is 3 seconds. The time from the moment the user clicks the button to the moment when the loading animation is hidden and the page element corresponding to the button is displayed is the operation duration information. For example, from the time the user clicks the button to the time the final consultation record is created and displayed to the user, it takes 8 seconds as a whole, so the operation duration information is 8 seconds.

[0069] It should be noted that this embodiment can also identify the reasons for long page or element wait times. For example, long page load times may be due to slow backend interface response; failure to load certain elements may be due to network issues or front-end code errors. Furthermore, based on the identification results, optimization suggestions are provided. For example, back-end interface performance optimization to reduce response time; front-end code optimization to reduce page loading resources; and asynchronous loading technology to optimize the user experience.

[0070] based on Figure 1 In the illustrated embodiment, step S20 further includes:

[0071] Determining an interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information;

[0072] In a case where the interface status information is the serial status information, taking the sum of the interface response times of all serial interfaces as the serial interface response time;

[0073] In a case where the interface state information is the parallel state information, determining the maximum interface response time among the parallel interfaces as the parallel interface response time;

[0074] The sum of the serial interface response time and / or the parallel interface response time is determined as the operation duration information.

[0075] In one embodiment, the calling sequence and dependency relationships of interface requests are analyzed to determine interface status information. If a request from interface B depends on a response from interface A, interfaces A and B are serially called. For example, interface A → interface B → interface C is a serial call. The response time of all serial interfaces is accumulated to obtain the serial interface response time.

[0076] Based on the interface status information, all interfaces called in parallel are identified. If multiple interfaces are initiated simultaneously and are independent of each other, they are parallel calls. If interfaces D, E, and F are initiated simultaneously and are independent of each other, then interfaces D, E, and F are parallel interfaces. The interface with the longest response time is taken as the parallel interface response time.

[0077] If there is only a serial interface or only a parallel interface in the current monitoring page, the operation duration information is the corresponding serial or parallel interface response time; if the current monitoring page has both a serial interface and a parallel interface, the serial interface response time and the parallel interface response time are added together to obtain the operation duration information.

[0078] In a more preferred embodiment, determining the interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information, includes:

[0079] Determining a network request sequence according to the user operation instruction;

[0080] According to the network request sequence, interface status information corresponding to the current monitoring page is determined.

[0081] In one embodiment, in the financial sector, user actions (such as login, transaction, query, etc.) typically trigger a series of network requests. For example, on a front-end page (such as a web or mobile application), user action instructions are captured through event monitoring. For example, a user clicks the "Submit Transaction" button or enters the transaction amount. A front-end performance monitoring tool is used to capture all network requests triggered by user actions and record detailed information for each request, including the request URL, request initiation time, request completion time, and request status code.

[0082] Sort the captured network requests according to the request initiation time, build a network request sequence to determine the order and dependency of each request, and determine the interface status information:

[0083] Serial status information: If request B depends on the completion of request A, it is marked as serial;

[0084] Parallel status information: If multiple requests are initiated simultaneously and are not dependent on each other, they are marked as parallel.

[0085] Based on any of the above embodiments, in this embodiment, step S10 includes:

[0086] Collect network request data through preset non-intrusive data collection methods;

[0087] The network request data in response to the user operation instruction is determined as the current monitoring page, and the element in the current monitoring page that matches the user operation instruction is determined as the current monitoring element.

[0088] Specifically, non-intrusive means that there is no need to modify the source code of the monitored or extended system, nor is there any need to introduce excessive dependencies or make large-scale adjustments to the system's runtime environment. Use non-intrusive data collection tools to capture all network requests triggered by user operations and record detailed information for each request.

[0089] Listen for user action events (such as clicking a button or submitting a form), capture the action instructions, determine the current monitored page based on the user action instructions, and find the element on the current page that matches the action instructions. For example, if a user clicks the "Submit Transaction" button, the current monitored page is the "Transaction Page." When the user clicks the "Submit Transaction" button, the matching current monitored element is the button itself.

[0090] See also Figure 3 , Figure 3 The embodiment of the present application provides a schematic block diagram of a network performance monitoring device, which is used to execute the aforementioned network performance monitoring method. The network performance monitoring device can be configured on a server.

[0091] like Figure 3 As shown, the network performance monitoring device 400 includes:

[0092] A monitoring target determination module 410 is configured to determine, when a network request is triggered by a user operation instruction, a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction;

[0093] The duration information determination module 420 is configured to monitor change information of the current monitoring page and the current monitoring element, and determine waiting duration information and operation duration information respectively according to the change information;

[0094] The network performance prediction data generation module 430 is used to generate network performance prediction data through a preset machine learning algorithm, the waiting time information and the operation time information.

[0095] Furthermore, the network performance prediction data generating module 430 includes:

[0096] A historical performance feature vector extraction unit, configured to obtain historical network performance information and extract historical performance feature vectors from the historical network performance information;

[0097] an abnormal fluctuation feature matrix construction unit, configured to determine network abnormal fluctuation information by using the preset machine learning algorithm, the waiting time information, and the operation time information, and to construct an abnormal fluctuation feature matrix of the network abnormal fluctuation information;

[0098] The network performance prediction data generating unit is used to generate the network performance prediction data according to the historical performance feature vector and the abnormal fluctuation feature matrix.

[0099] Furthermore, the network performance prediction data generating unit includes:

[0100] a comprehensive feature matrix generating subunit, configured to fuse the historical performance feature vector with the abnormal fluctuation feature matrix to generate a comprehensive feature matrix;

[0101] The network performance prediction data generating subunit is used to recursively predict the comprehensive feature matrix through the preset machine learning algorithm to generate the network performance prediction data.

[0102] Furthermore, the duration information determination module 420 includes:

[0103] a preset loading element determining unit, configured to determine whether a preset loading element exists on the current monitoring page according to change information of the current monitoring element;

[0104] The waiting time information determining unit is used to determine the waiting time information when the preset loading element exists on the current monitoring page.

[0105] Furthermore, the duration information determination module 420 includes:

[0106] an interface information determining unit, configured to determine an interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information;

[0107] a serial interface response time determining unit, configured to, when the interface state information is the serial state information, take the sum of the interface response times of all serial interfaces as the serial interface response time;

[0108] a parallel interface response time determining unit, configured to determine, when the interface state information is the parallel state information, a maximum interface response time among the parallel interfaces as the parallel interface response time;

[0109] The operation duration information determining unit is configured to determine the sum of the serial interface response time and / or the parallel interface response time as the operation duration information.

[0110] Furthermore, the interface information determining unit includes:

[0111] A network request sequence determining subunit, configured to determine a network request sequence according to the user operation instruction;

[0112] The interface status information determining subunit is used to determine the interface status information corresponding to the current monitoring page according to the network request sequence.

[0113] Furthermore, the monitoring target determination module 410 includes:

[0114] A network request data collection unit, configured to collect network request data using a preset non-intrusive data collection method;

[0115] A monitoring target determination unit is configured to determine the network request data in response to the user operation instruction as the current monitoring page, and determine the element in the current monitoring page that matches the user operation instruction as the current monitoring element.

[0116] It should be noted that those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] The above-mentioned device can be realized in the form of a computer program. The computer program can be used in Figure 4 Runs on the computer device shown.

[0118] See also Figure 4 , Figure 4 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0119] See Figure 4The computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0120] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any network performance monitoring method.

[0121] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0122] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any network performance monitoring method.

[0123] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0124] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0125] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0126] In the case where a network request is triggered by a user operation instruction, determining a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction;

[0127] Monitor change information of the current monitoring page and the current monitoring element, and determine waiting time information and operation time information respectively according to the change information;

[0128] Network performance prediction data is generated by using a preset machine learning algorithm, the waiting time information, and the operation time information.

[0129] In one embodiment, network performance prediction data is generated by using a preset machine learning algorithm, the waiting time information, and the operation time information to achieve:

[0130] Acquiring historical network performance information, and extracting historical performance feature vectors from the historical network performance information;

[0131] Determine network abnormal fluctuation information through the preset machine learning algorithm, the waiting time information, and the operation time information, and construct an abnormal fluctuation feature matrix of the network abnormal fluctuation information;

[0132] The network performance prediction data is generated according to the historical performance feature vector and the abnormal fluctuation feature matrix.

[0133] In one embodiment, the network performance prediction data is generated based on the historical performance feature vector and the abnormal fluctuation feature matrix to achieve:

[0134] Fusing the historical performance feature vector with the abnormal fluctuation feature matrix to generate a comprehensive feature matrix;

[0135] The comprehensive feature matrix is ​​recursively predicted by the preset machine learning algorithm to generate the network performance prediction data.

[0136] In one embodiment, the change information of the currently monitored page and the currently monitored element is monitored, and the waiting time information and the operation time information are determined respectively according to the change information, so as to achieve:

[0137] Determining whether a preset loading element exists on the current monitoring page according to the change information of the current monitoring element;

[0138] In a case where the preset loading element exists on the current monitoring page, the waiting time information is determined.

[0139] In one embodiment, the change information of the currently monitored page and the currently monitored element is monitored, and the waiting time information and the operation time information are determined respectively according to the change information, so as to achieve:

[0140] Determining an interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information;

[0141] In a case where the interface status information is the serial status information, taking the sum of the interface response times of all serial interfaces as the serial interface response time;

[0142] In a case where the interface state information is the parallel state information, determining the maximum interface response time among the parallel interfaces as the parallel interface response time;

[0143] The sum of the serial interface response time and / or the parallel interface response time is determined as the operation duration information.

[0144] In one embodiment, determining an interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information, is used to implement:

[0145] Determining a network request sequence according to the user operation instruction;

[0146] According to the network request sequence, interface status information corresponding to the current monitoring page is determined.

[0147] In one embodiment, when a network request is triggered by a user operation instruction, the current monitoring page and the current monitoring element in the current monitoring page are determined according to the user operation instruction to achieve:

[0148] Collect network request data through preset non-intrusive data collection methods;

[0149] The network request data in response to the user operation instruction is determined as the current monitoring page, and the element in the current monitoring page that matches the user operation instruction is determined as the current monitoring element.

[0150] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and the computer program includes program instructions. The processor executes the program instructions to implement any network performance monitoring method provided in the embodiment of the present application.

[0151] The computer-readable storage medium may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a flash memory card, etc., equipped on the computer device.

[0152] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A network performance monitoring method, characterized in that: include: In the case where a network request is triggered by a user operation instruction, determining a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction; Monitor change information of the current monitoring page and the current monitoring element, and determine waiting time information and operation time information respectively according to the change information; Network performance prediction data is generated by using a preset machine learning algorithm, the waiting time information, and the operation time information.

2. The network performance monitoring method according to claim 1, wherein: The generating of network performance prediction data by using a preset machine learning algorithm, the waiting time information, and the operation time information includes: Acquiring historical network performance information, and extracting historical performance feature vectors from the historical network performance information; Determine network abnormal fluctuation information through the preset machine learning algorithm, the waiting time information, and the operation time information, and construct an abnormal fluctuation feature matrix of the network abnormal fluctuation information; The network performance prediction data is generated according to the historical performance feature vector and the abnormal fluctuation feature matrix.

3. The network performance monitoring method according to claim 2, wherein: The generating the network performance prediction data according to the historical performance characteristic vector and the abnormal fluctuation characteristic matrix includes: Fusing the historical performance feature vector with the abnormal fluctuation feature matrix to generate a comprehensive feature matrix; The comprehensive feature matrix is ​​recursively predicted by the preset machine learning algorithm to generate the network performance prediction data.

4. The network performance monitoring method according to claim 1, wherein: The monitoring of change information of the current monitoring page and the current monitoring element, and determining waiting time information and operation time information respectively according to the change information, includes: Determining whether a preset loading element exists on the current monitoring page according to the change information of the current monitoring element; In a case where the preset loading element exists on the current monitoring page, the waiting time information is determined.

5. The network performance monitoring method according to claim 4, characterized in that: The monitoring of change information of the current monitoring page and the current monitoring element, and determining waiting time information and operation time information respectively according to the change information, includes: Determining an interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information; In a case where the interface status information is the serial status information, taking the sum of the interface response times of all serial interfaces as the serial interface response time; In a case where the interface state information is the parallel state information, determining the maximum interface response time among the parallel interfaces as the parallel interface response time; The sum of the serial interface response time and / or the parallel interface response time is determined as the operation duration information.

6. The network performance monitoring method according to claim 5, characterized in that: The determining of the interface response time and interface status information corresponding to the current monitoring page, wherein the interface status information includes at least one of serial status information and parallel status information, includes: Determining a network request sequence according to the user operation instruction; According to the network request sequence, interface status information corresponding to the current monitoring page is determined.

7. The network performance monitoring method according to any one of claims 1 to 6, characterized in that: In the case where the network request is triggered by the user operation instruction, determining the current monitoring page and the current monitoring element in the current monitoring page according to the user operation instruction includes: Collect network request data through preset non-intrusive data collection methods; The network request data in response to the user operation instruction is determined as the current monitoring page, and the element in the current monitoring page that matches the user operation instruction is determined as the current monitoring element.

8. A network performance monitoring device, characterized in that: include: A monitoring target determination module is used to determine a current monitoring page and a current monitoring element in the current monitoring page according to the user operation instruction when a network request is triggered by the user operation instruction; a duration information determination module, configured to monitor change information of the currently monitored page and the currently monitored element, and determine waiting duration information and operation duration information respectively according to the change information; The network performance prediction data generation module is used to generate network performance prediction data through a preset machine learning algorithm, the waiting time information and the operation time information.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the network performance monitoring method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to implement the network performance monitoring method according to any one of claims 1 to 7.

Images