A business operation monitoring method and system based on microservices

Abstract

The present invention relates to a business operation monitoring method based on microservices, and to the field of data processing systems or methods specifically suitable for supervision or prediction purposes, the method comprising: performing multiple training on a convolutional neural network to obtain a convolutional neural network model; using the convolutional neural network model to intelligently predict a single faulty microservice data corresponding to the overall system in the current time interval based on multiple faulty microservice data corresponding to the overall system in multiple past time intervals and the deployment information of various microservices deployed by the overall system. The present invention also relates to a business operation monitoring system based on microservices. Through the present invention, in view of the technical problem that the state of microservices in the future cannot be predicted, a convolutional neural network model with a targeted structural design is introduced to intelligently predict the fault information of microservices in the future based on the data processing results of microservice logs and the deployment information of microservices, thereby solving the above technical problems.

Description

Technical Field

[0001] The present invention relates to the field of data processing systems or methods specifically suitable for monitoring or prediction purposes, and in particular to a business operation monitoring method and system based on microservices. Background Art

[0002] Microservices is a cloud-native architectural approach that includes many small, loosely coupled and independently deployable components or services in a single application. These services typically have their own technology stack, including databases and data management models; communicate with each other through a combination of REST APIs, event streams, and message brokers; and are organized according to business capabilities, with service delimiters usually called bounded contexts. Microservices are characterized by easier code updates: new features or functions can be added directly without having to update the entire application, and teams can use different technology stacks and different programming languages ​​for different components. Components can be scaled independently of each other, reducing the waste and cost associated with having to scale the entire application. The difference between microservices and monolithic architectures is that microservices consist of many smaller, loosely coupled services that make up an application, as opposed to the monolithic approach of large, tightly coupled applications. In a system with multiple microservices deployed, the operating status of the microservices, such as whether they are in a fault state, is the main focus and supervision of the system.

[0003] For example, the Chinese invention patent with publication number CN111694721A proposes a microservice fault monitoring method and device, including: obtaining input parameters; extracting the log information of the microservice from a preset message queue according to the input parameters; the log information includes multiple abnormal log information; using the multiple abnormal log information to generate a fault project library; assigning corresponding weights to the microservice; generating a frequent fault project set according to the weight and the fault project library; using the frequent fault project set to build a fault tree, and displaying the fault tree in a preset monitoring interface. The embodiment of the present invention combines the weight of the microservice to perform correlation monitoring on the frequent fault items, avoiding the problem that the frequent item correlation monitoring process occupies a large amount of memory space, resulting in a long time-consuming fault tree construction, so that a fault tree can be quickly built from the log when a fault occurs in the power metering system to achieve fault location.

[0004] For example, the Chinese invention patent with publication number CN110888783A proposes a monitoring method, device and electronic device for a microservice system, which relates to the field of computer technology. The method includes: obtaining monitoring data of at least one microservice node in the microservice system through a preset business interception model; generating time series data of each microservice node based on the monitoring data of each microservice node, and storing the time series data; if it is detected that the time series data of any stored microservice node meets the data abnormality condition, sending corresponding data abnormality prompt information.

[0005] It can be seen that in the prior art, the monitoring schemes for faulty microservices in a system with multiple microservices deployed are all based on the real-time monitoring data or past monitoring data of each microservice in the past, such as the current time log or the past time log of the microservice. The analysis operation of the microservices that have failed at the current time or in the past time is performed on the system with multiple microservices deployed, and the microservices that have failed at the future time cannot be predicted, which makes it difficult to realize the business operation monitoring of microservices in the future time, and further, it is impossible to respond in advance and solve the faults of microservices that will fail in the future time, which makes the robustness of the entire system lack further room for improvement. Summary of the invention

[0006] In order to solve the technical defects in the prior art, the present invention provides a business operation monitoring method and system based on microservices, which can perform customized data analysis on key logs of each past time interval of a system deployed with multiple microservices, and collect microservice deployment information of the system deployed with multiple microservices, so as to provide comprehensive and sufficient reference data for the prediction of faulty microservices in future time intervals of the system, and at the same time, a convolutional neural network model for executing the prediction of faulty microservices in future time intervals is designed with a targeted structure, so as to complete the reliable prediction of faulty microservices in future time intervals, realize business operation monitoring of microservices in the future, and provide a processing solution for early response and fault resolution of microservices that fail in the future and improve the robustness of the entire system.

[0007] According to a first aspect of the present invention, a microservice-based business operation monitoring method is provided, the method comprising:

[0008] Taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals;

[0009] Processing a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed;

[0010] Obtaining deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system includes a type name, a version number, and a system number of the information system to which each microservice deployed by the overall system is deployed;

[0011] Performing multiple training on the convolutional neural network to obtain a convolutional neural network after completing multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system;

[0012] The convolutional neural network model is used to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems;

[0013] Among them, the single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed.

[0014] According to a second aspect of the present invention, a microservice-based business operation monitoring system is provided, the system comprising:

[0015] The log operation mechanism regards multiple information systems with data interaction as an overall system, collects multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performs data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals;

[0016] a data analysis mechanism connected to the log operation mechanism, and used to process information of a plurality of key logs corresponding to the overall system in a plurality of past time intervals, so as to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed;

[0017] A deployment monitoring mechanism is used to obtain deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system is the type name, version number and system number of the information system to which each microservice deployed by the overall system is deployed;

[0018] A multiple training mechanism, used to perform multiple training on the convolutional neural network to obtain a convolutional neural network after completing multiple training and output it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system;

[0019] A prediction execution mechanism, connected to the log operation mechanism, the data analysis mechanism, the deployment monitoring mechanism and the multiple training mechanism, respectively, for using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system and the number of systems of the multiple information systems;

[0020] Among them, the single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed.

[0021] It can be seen that the present invention has at least the following five main inventive points:

[0022] First, multiple information systems with data interaction are regarded as an overall system, and data analysis of multiple microservice call log sets corresponding to multiple past time intervals is performed on the overall system to obtain multiple key logs corresponding to the overall system in multiple past time intervals. Specifically, the last log that appears continuously in the single microservice call log set corresponding to the overall system in a certain past time interval and calls the same microservice is used as the single key log corresponding to the overall system in the certain past time interval, thereby completing the targeted screening of the key logs in each past time interval and providing a data basis for the subsequent prediction of faulty microservices;

[0023] Second: performing information processing on multiple key logs corresponding to the overall system in multiple past time intervals to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals, each fault microservice data including the type name of the faulty microservice, the version number of the faulty microservice and the system number of the information system to which the faulty microservice is deployed. Specifically, performing data analysis on a single key log corresponding to the overall system in a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, identifying the corresponding microservice as a microservice of the overall system that has a fault in the past time interval to determine the single fault microservice data corresponding to the overall system in the past time interval, thereby completing customized judgment of microservices in a faulty state based on a single key log, and further providing a data basis for subsequent prediction of faulty microservices;

[0024] Third: A convolutional neural network model with targeted structural design is used to intelligently predict a single faulty microservice data corresponding to the overall system in the current time interval based on multiple faulty microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of multiple information systems, thereby completing the intelligent prediction of a single faulty microservice data corresponding to the overall system in the current time interval based on the faulty microservice data of multiple past time intervals of the overall system and the deployment information of various microservices deployed by the overall system, providing a predictive automatic monitoring mechanism for accurate judgment and effective identification of future microservice fault data analysis based on microservice deployment status monitoring results, microservice version information monitoring results, and past microservice fault data analysis results;

[0025] Fourthly, the convolutional neural network is trained multiple times to obtain a convolutional neural network after multiple training and output as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed in the overall system, thereby building convolutional neural network models with different structures for different overall systems, and providing a guarantee for the reliability and stability of the prediction results of the convolutional neural network;

[0026] Fifth: In each training performed on the convolutional neural network, a single known fault microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system and the number of systems of the multiple information systems are used as the input content of the convolutional neural network to complete this training of the convolutional neural network, thereby improving the training effect of each training of the convolutional neural network, and further providing guarantee for the reliability and stability of the prediction results of the convolutional neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0028] Figure 1 It is a technical flow chart of the microservice-based business operation monitoring method and system according to the present invention.

[0029] Figure 2 The present invention is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 1 of the present invention.

[0030] Figure 3 The present invention is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 2 of the present invention.

[0031] Figure 4 This is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 3 of the present invention.

[0032] Figure 5 This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 4 of the present invention.

[0033] Figure 6 This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 5 of the present invention.

[0034] Figure 7 This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 6 of the present invention. DETAILED DESCRIPTION

[0035] like Figure 1 As shown, a technical flow chart of a microservice-based business operation monitoring method and system according to the present invention is given.

[0036] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0037] Technical process 1: multiple information systems with data interaction are regarded as an overall system, and data analysis of multiple microservice call log sets corresponding to multiple past time intervals is performed on the overall system to obtain multiple key logs corresponding to the overall system in multiple past time intervals, and information processing is performed on the multiple key logs to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals;

[0038] like Figure 1 As shown in the figure, an overall system consisting of 6 information systems with data interaction is given. Obviously, the 6 information systems are only exemplary and are not limited to the actual application. Figure 1 The 6 information systems shown;

[0039] In this way, by processing the data of microservices in multiple past time intervals of the overall system, the first basic data for monitoring or predicting the business operation status of microservices is obtained. The business operation status of microservices includes whether the microservice is in a fault state;

[0040] Technical process 2: obtaining deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system includes the type name, version number and system number of the information system to which each microservice deployed by the overall system is deployed;

[0041] In this way, the deployment information of various microservices deployed by the overall system is used as the second basic data for monitoring or predicting the business operation status of microservices;

[0042] Technical process three: Targeted design of a convolutional neural network model for predicting the future microservice business operation status. The design of the convolutional neural network model is targeted in the following aspects:

[0043] First, the convolutional neural network model is a convolutional neural network that has completed multiple trainings. The number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed in the overall system, thereby constructing convolutional neural network models with different structures for different overall systems, thereby providing a guarantee for the reliability and stability of the prediction results of the convolutional neural network;

[0044] Second, in each training of the convolutional neural network, a single piece of faulty microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple pieces of faulty microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems are used as the input content of the convolutional neural network to complete this training of the convolutional neural network;

[0045] Through the design of the above training mechanism, the training effect of each training of the convolutional neural network is improved, and the reliability and stability of the prediction results of the convolutional neural network are further guaranteed;

[0046] Technical process 4: The convolutional neural network model designed in technical process 3 is used to predict the fault microservice data of the entire system in the future time interval based on multiple basic data obtained from technical processes 1 and 2;

[0047] Specifically, the future time interval is the current time interval, and the current time interval takes the current moment as the starting moment, and is therefore a future time interval;

[0048] For example, the predicted faulty microservice data of the entire system in a future time interval includes the type name of the microservice that fails in the entire system in the future time interval, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed;

[0049] In this way, the predicted type names of microservices that will fail in the future time interval of the entire system, the version numbers of the failed microservices, and the system numbers of the information systems to which the failed microservices are deployed can be sent to a monitoring and display agency for on-site monitoring. This can also provide valuable reference content and solution opportunities for advance response and troubleshooting of microservices that may fail in the future and for improving the robustness of the entire system.

[0050] The key points of the present invention are: customized analysis of microservice log data in past time intervals and targeted selection of microservice deployment information to obtain sufficient and comprehensive basic data for predicting faulty microservices in the future, and targeted structural design of multiple aspects of a convolutional neural network model for performing prediction of faulty microservices in the future.

[0051] The microservice-based business operation monitoring method and system of the present invention will be specifically described below by way of embodiments.

[0052] Example 1

[0053] Figure 2 The present invention is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 1 of the present invention.

[0054] like Figure 2 As shown, the microservice-based business operation monitoring method includes the following steps:

[0055] Step S101: taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals;

[0056] For example, multiple information systems with data interaction are regarded as an overall system, multiple microservice call log sets corresponding to the overall system in multiple past time intervals are collected, and data analysis is performed to obtain multiple key logs corresponding to the overall system in multiple past time intervals. The method includes: when a data communication link is established between two information systems, the two information systems can be regarded as two information systems with data interaction;

[0057] Step S102: performing information processing on a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed;

[0058] Specifically, information processing is performed on a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed. Even if it is the same microservice, its version number is different due to the sequence of research and development;

[0059] Step S103: acquiring deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system includes the type name, version number and system number of the information system to which each microservice deployed by the overall system is deployed;

[0060] Step S104: performing multiple training on the convolutional neural network to obtain a convolutional neural network after multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system;

[0061] For example, performing multiple training on a convolutional neural network to obtain a convolutional neural network after multiple training and outputting it as a convolutional neural network model, the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: the total number of various microservices deployed by the overall system is 20, the number of trainings performed by the convolutional neural network is 100, the total number of various microservices deployed by the overall system is 30, the number of trainings performed by the convolutional neural network is 200, the total number of various microservices deployed by the overall system is 40, the number of trainings performed by the convolutional neural network is 300, the total number of various microservices deployed by the overall system is 50, the number of trainings performed by the convolutional neural network is 400, and so on;

[0062] Step S105: using the convolutional neural network model to intelligently predict a single piece of faulty microservice data corresponding to the overall system in the current time interval based on multiple pieces of faulty microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems;

[0063] The single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed;

[0064] Among them, the convolutional neural network is trained multiple times to obtain the convolutional neural network after completing the multiple trainings and output as the convolutional neural network model, and the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: in each training performed on the convolutional neural network, a single known fault microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems are used as the input content of the convolutional neural network to complete the current training of the convolutional neural network;

[0065] Specifically, in each training performed on the convolutional neural network, a single piece of fault microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple pieces of fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems are used as the input content of the convolutional neural network. The training of the convolutional neural network includes: each training performed on the convolutional neural network can be completed by using the MATLAB toolbox;

[0066] Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals includes: taking the last log that appears continuously in the single microservice call log set corresponding to the overall system in a certain past time interval and calls the same microservice as the single key log corresponding to the overall system in the certain past time interval;

[0067] And wherein, information processing is performed on multiple key logs corresponding to the overall system in multiple past time intervals to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals, each fault microservice data includes the type name of the faulty microservice, the version number of the faulty microservice and the system number of the information system to which the faulty microservice is deployed, including: data analysis is performed on a single key log corresponding to the overall system in a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that failed in the past time interval to determine the single fault microservice data corresponding to the overall system in the past time interval.

[0068] Example 2

[0069] Figure 3 The present invention is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 2 of the present invention.

[0070] like Figure 3As shown, after using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems, that is, after step S105, the microservice-based business operation monitoring method further includes:

[0071] Step S106: receiving and displaying in real time the type name of the microservice that fails in the current time interval, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed;

[0072] For example, an LED display array, an LCD display array, or a giant display screen may be selected to receive and display in real time the type name of the microservice that fails in the current time interval of the overall system, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed.

[0073] Example 3

[0074] Figure 4 This is a flowchart of the steps of a microservice-based business operation monitoring method according to Embodiment 3 of the present invention.

[0075] like Figure 4 As shown, after performing information processing on a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including the type name of the faulty microservice, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed, that is, after step S102, the microservice-based business operation monitoring method further includes:

[0076] Step S107: storing a plurality of faulty microservice data corresponding to the overall system in a plurality of past time intervals in a database in a database storage unit;

[0077] For example, storing a plurality of faulty microservice data corresponding to the overall system in a plurality of past time intervals in a database in a database storage unit includes: the type of the database is an Oracle database or an ODB database.

[0078] In any of the above embodiments 1-3, optionally, in the microservice-based business operation monitoring method:

[0079] Taking multiple information systems with data interaction as an integral system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals, further comprising: multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration;

[0080] For example, the multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration, including: the set duration is one of 10 minutes, 15 minutes or 30 minutes;

[0081] Wherein, a data analysis is performed on a single key log corresponding to the overall system within a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that has failed within the certain past time interval to determine the single fault microservice data corresponding to the overall system within the certain past time interval; the single fault microservice data corresponding to the overall system within the certain past time interval includes the type name of the microservice of the overall system that has failed within the certain past time interval, the version number of the microservice of the overall system that has failed within the certain past time interval, and the system number of the information system to which the microservice of the overall system that has failed within the certain past time interval is deployed;

[0082] Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals also includes: the current time interval takes the current time as the starting time;

[0083] Among them, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: synchronously inputting the multiple fault microservice data corresponding to the overall system in multiple past time intervals, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems into the convolutional neural network model;

[0084] And wherein, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: running the convolutional neural network model to obtain the single fault microservice data corresponding to the overall system in the current time interval output by the convolutional neural network model.

[0085] Example 4

[0086] Figure 5 This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 4 of the present invention.

[0087] like Figure 5 As shown, the microservice-based business operation monitoring system includes the following components:

[0088] The log operation mechanism regards multiple information systems with data interaction as an overall system, collects multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performs data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals;

[0089] For example, multiple information systems with data interaction are regarded as an overall system, multiple microservice call log sets corresponding to the overall system in multiple past time intervals are collected, and data analysis is performed to obtain multiple key logs corresponding to the overall system in multiple past time intervals. The method includes: when a data communication link is established between two information systems, the two information systems can be regarded as two information systems with data interaction;

[0090] a data analysis mechanism connected to the log operation mechanism, and used to process information of a plurality of key logs corresponding to the overall system in a plurality of past time intervals, so as to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed;

[0091] Specifically, information processing is performed on a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed. Even if it is the same microservice, its version number is different due to the sequence of research and development;

[0092] A deployment monitoring mechanism is used to obtain deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system is the type name, version number and system number of the information system to which each microservice deployed by the overall system is deployed;

[0093] A multiple training mechanism, used to perform multiple training on the convolutional neural network to obtain a convolutional neural network after completing multiple training and output it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system;

[0094] For example, performing multiple training on a convolutional neural network to obtain a convolutional neural network after multiple training and outputting it as a convolutional neural network model, the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: the total number of various microservices deployed by the overall system is 20, the number of trainings performed by the convolutional neural network is 100, the total number of various microservices deployed by the overall system is 30, the number of trainings performed by the convolutional neural network is 200, the total number of various microservices deployed by the overall system is 40, the number of trainings performed by the convolutional neural network is 300, the total number of various microservices deployed by the overall system is 50, the number of trainings performed by the convolutional neural network is 400, and so on;

[0095] A prediction execution mechanism, connected to the log operation mechanism, the data analysis mechanism, the deployment monitoring mechanism and the multiple training mechanism, respectively, for using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system and the number of systems of the multiple information systems;

[0096] The single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed;

[0097] Among them, the convolutional neural network is trained multiple times to obtain the convolutional neural network after completing the multiple trainings and output as the convolutional neural network model, and the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: in each training performed on the convolutional neural network, a single known fault microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems are used as the input content of the convolutional neural network to complete the current training of the convolutional neural network;

[0098] Specifically, in each training performed on the convolutional neural network, a single piece of fault microservice data corresponding to the overall system in a certain past time interval is used as the output content of the convolutional neural network, and multiple pieces of fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems are used as the input content of the convolutional neural network. The training of the convolutional neural network includes: each training performed on the convolutional neural network can be completed by using the MATLAB toolbox;

[0099] Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals includes: taking the last log that appears continuously in the single microservice call log set corresponding to the overall system in a certain past time interval and calls the same microservice as the single key log corresponding to the overall system in the certain past time interval;

[0100] And wherein, information processing is performed on multiple key logs corresponding to the overall system in multiple past time intervals to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals, each fault microservice data includes the type name of the faulty microservice, the version number of the faulty microservice and the system number of the information system to which the faulty microservice is deployed, including: data analysis is performed on a single key log corresponding to the overall system in a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that failed in the past time interval to determine the single fault microservice data corresponding to the overall system in the past time interval.

[0101] Example 5

[0102] Figure 6 This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 5 of the present invention.

[0103] like Figure 6 As shown, the microservice-based business operation monitoring system also includes:

[0104] A monitoring and display mechanism, connected to the prediction execution mechanism, for receiving and displaying in real time the type name of the microservice that fails in the overall system within the current time interval, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed;

[0105] For example, an LED display array, an LCD display array or a giant display screen can be selected to implement the monitoring display mechanism, which is used to receive and display in real time the type name of the microservice that fails in the overall system within the current time interval, the version number of the failed microservice, and the system number of the information system to which the failed microservice is deployed.

[0106] Example 6

[0107] Figure 7This is an internal structure diagram of a microservice-based business operation monitoring system according to Embodiment 6 of the present invention.

[0108] like Figure 7 As shown, the microservice-based business operation monitoring system also includes:

[0109] A content storage mechanism, connected to the data analysis mechanism and having a built-in database storage unit, is used to store a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals in a database in the database storage unit;

[0110] For example, storing a plurality of faulty microservice data corresponding to the overall system in a plurality of past time intervals in a database in a database storage unit includes: the type of the database is an Oracle database or an ODB database.

[0111] And in any of the above embodiments 4-6, optionally, in the microservice-based business operation monitoring system:

[0112] Taking multiple information systems with data interaction as an integral system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals, further comprising: multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration;

[0113] For example, the multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration, including: the set duration is one of 10 minutes, 15 minutes or 30 minutes;

[0114] Wherein, a data analysis is performed on a single key log corresponding to the overall system within a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that has failed within the certain past time interval to determine the single fault microservice data corresponding to the overall system within the certain past time interval; the single fault microservice data corresponding to the overall system within the certain past time interval includes the type name of the microservice of the overall system that has failed within the certain past time interval, the version number of the microservice of the overall system that has failed within the certain past time interval, and the system number of the information system to which the microservice of the overall system that has failed within the certain past time interval is deployed;

[0115] Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals also includes: the current time interval takes the current time as the starting time;

[0116] Among them, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: synchronously inputting the multiple fault microservice data corresponding to the overall system in multiple past time intervals, the deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems into the convolutional neural network model;

[0117] And wherein, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: running the convolutional neural network model to obtain the single fault microservice data corresponding to the overall system in the current time interval output by the convolutional neural network model.

[0118] In addition, in the microservice-based business operation monitoring method and system according to the present invention:

[0119] Performing multiple training on the convolutional neural network to obtain a convolutional neural network after multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: using an information conversion function to represent an information conversion relationship of the positive correlation between the number of trainings performed by the convolutional neural network and the total number of various microservices deployed by the overall system;

[0120] For example, a CPLD chip may be selected to complete the testing and simulation of a data processing process that uses an information conversion function to represent a positive correlation between the number of trainings performed by the convolutional neural network and the total number of various microservices deployed by the overall system;

[0121] Specifically, the CPLD chip can be selected to complete the test and simulation of the data processing process of using the information conversion function to represent the positive correlation between the number of trainings performed by the convolutional neural network and the total number of various microservices deployed by the overall system, including: the CPLD chip is designed and implemented using the VHDL language;

[0122] The information conversion relationship of using an information conversion function to represent the positive correlation between the number of trainings performed by the convolutional neural network and the total number of various microservices deployed by the overall system includes: in the information conversion function, the total number of various microservices deployed by the overall system is used as input information of the information conversion function;

[0123] And wherein, the information conversion relationship that uses an information conversion function to represent the positive correlation between the number of training times performed by the convolutional neural network and the total number of various microservices deployed by the overall system also includes: in the information conversion function, the number of training times performed by the convolutional neural network is used as the output information of the information conversion function.

[0124] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the relevant technology, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A microservice-based business operation monitoring method, characterized in that: The method comprises: Taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals; Processing a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed; Obtaining deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system includes a type name, a version number, and a system number of the information system to which each microservice deployed by the overall system is deployed; Performing multiple training on the convolutional neural network to obtain a convolutional neural network after completing multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system; The convolutional neural network model is used to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems; Among them, the single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed.

2. The microservice-based business operation monitoring method according to claim 1, characterized in that: Performing multiple training on the convolutional neural network to obtain the convolutional neural network after completing multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: in each training performed on the convolutional neural network, using a known single fault microservice data corresponding to the overall system in a certain past time interval as the output content of the convolutional neural network, using multiple fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems as the input content of the convolutional neural network to complete this training of the convolutional neural network; Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals includes: taking the last log that appears continuously in the single microservice call log set corresponding to the overall system in a certain past time interval and calls the same microservice as the single key log corresponding to the overall system in the certain past time interval; Among them, information processing is performed on multiple key logs corresponding to the overall system in multiple past time intervals to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals, each fault microservice data includes the type name of the faulty microservice, the version number of the faulty microservice and the system number of the information system to which the faulty microservice is deployed, including: data analysis is performed on a single key log corresponding to the overall system in a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that failed in the past time interval to determine the single fault microservice data corresponding to the overall system in the past time interval.

3. The microservice-based business operation monitoring method according to claim 2, characterized in that: After using the convolutional neural network model to intelligently predict a single copy of faulty microservice data corresponding to the overall system in the current time interval based on multiple copies of faulty microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems, the method further includes: Receive and display in real time the type name of the microservice that fails in the current time interval of the overall system, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed.

4. The microservice-based business operation monitoring method according to claim 2, characterized in that: After performing information processing on a plurality of key logs corresponding to the overall system in a plurality of past time intervals to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed, the method further includes: A plurality of faulty microservice data corresponding to the overall system in a plurality of past time intervals are stored in a database in a database storage unit.

5. The microservice-based business operation monitoring method according to any one of claims 2 to 4, characterized in that: Taking multiple information systems with data interaction as an integral system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals, further comprising: multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration; Wherein, a data analysis is performed on a single key log corresponding to the overall system within a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that has failed within the certain past time interval to determine the single fault microservice data corresponding to the overall system within the certain past time interval; the single fault microservice data corresponding to the overall system within the certain past time interval includes the type name of the microservice of the overall system that has failed within the certain past time interval, the version number of the microservice of the overall system that has failed within the certain past time interval, and the system number of the information system to which the microservice of the overall system that has failed within the certain past time interval is deployed; Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals also includes: the current time interval takes the current time as the starting time; Among them, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple copies of fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: synchronously inputting multiple copies of fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems into the convolutional neural network model; running the convolutional neural network model to obtain a single copy of fault microservice data corresponding to the overall system in the current time interval output by the convolutional neural network model.

6. A microservice-based business operation monitoring system, characterized in that: The system comprises: The log operation mechanism regards multiple information systems with data interaction as an overall system, collects multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performs data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals; a data analysis mechanism connected to the log operation mechanism, and used to process information of a plurality of key logs corresponding to the overall system in a plurality of past time intervals, so as to obtain a plurality of fault microservice data corresponding to the overall system in a plurality of past time intervals, each piece of fault microservice data including a type name of a faulty microservice, a version number of a faulty microservice, and a system number of an information system to which the faulty microservice is deployed; A deployment monitoring mechanism is used to obtain deployment information of various microservices deployed by the overall system, wherein the deployment information of each microservice deployed by the overall system is the type name, version number and system number of the information system to which each microservice deployed by the overall system is deployed; A multiple training mechanism, used to perform multiple training on the convolutional neural network to obtain a convolutional neural network after completing multiple training and output it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system; A prediction execution mechanism, connected to the log operation mechanism, the data analysis mechanism, the deployment monitoring mechanism and the multiple training mechanism, respectively, for using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system and the number of systems of the multiple information systems; Among them, the single faulty microservice data corresponding to the overall system in the current time interval includes the type name of the microservice that fails in the overall system in the current time interval, the version number of the faulty microservice, and the system number of the information system to which the faulty microservice is deployed.

7. The microservice-based business operation monitoring system according to claim 6, characterized in that: Performing multiple training on the convolutional neural network to obtain the convolutional neural network after completing multiple training and outputting it as a convolutional neural network model, wherein the number of trainings performed by the convolutional neural network is positively correlated with the total number of various microservices deployed by the overall system, including: in each training performed on the convolutional neural network, using a known single fault microservice data corresponding to the overall system in a certain past time interval as the output content of the convolutional neural network, using multiple fault microservice data corresponding to the overall system in multiple time intervals before the certain past time interval, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems as the input content of the convolutional neural network to complete this training of the convolutional neural network; Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals includes: taking the last log that appears continuously in the single microservice call log set corresponding to the overall system in a certain past time interval and calls the same microservice as the single key log corresponding to the overall system in the certain past time interval; Among them, information processing is performed on multiple key logs corresponding to the overall system in multiple past time intervals to obtain multiple fault microservice data corresponding to the overall system in multiple past time intervals, each fault microservice data includes the type name of the faulty microservice, the version number of the faulty microservice and the system number of the information system to which the faulty microservice is deployed, including: data analysis is performed on a single key log corresponding to the overall system in a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that failed in the past time interval to determine the single fault microservice data corresponding to the overall system in the past time interval.

8. The microservice-based business operation monitoring system according to claim 7, characterized in that: The system further comprises: The monitoring and display mechanism is connected to the prediction execution mechanism, and is used to receive and display in real time the type name of the microservice that fails in the overall system within the current time interval, the version number of the microservice that fails, and the system number of the information system to which the microservice that fails is deployed.

9. The microservice-based business operation monitoring system according to claim 7, characterized in that: The system further comprises: The content storage mechanism is connected to the data analysis mechanism and has a built-in database storage unit, and is used to store multiple fault microservice data corresponding to the overall system in multiple past time intervals in a database in the database storage unit.

10. The microservice-based business operation monitoring system according to any one of claims 7 to 9, characterized in that: Taking multiple information systems with data interaction as an integral system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals, further comprising: multiple past time intervals are between the current time interval and form a complete time segment with the current time interval, and the duration of each time interval in the multiple past time intervals and the current time interval is the set duration; Wherein, a data analysis is performed on a single key log corresponding to the overall system within a certain past time interval, and when it is determined that the microservice corresponding to the single key log is still in an unresolved fault state, the corresponding microservice is identified as a microservice of the overall system that has failed within the certain past time interval to determine the single fault microservice data corresponding to the overall system within the certain past time interval; the single fault microservice data corresponding to the overall system within the certain past time interval includes the type name of the microservice of the overall system that has failed within the certain past time interval, the version number of the microservice of the overall system that has failed within the certain past time interval, and the system number of the information system to which the microservice of the overall system that has failed within the certain past time interval is deployed; Among them, taking multiple information systems with data interaction as an overall system, collecting multiple microservice call log sets corresponding to the overall system in multiple past time intervals, and performing data analysis to obtain multiple key logs corresponding to the overall system in multiple past time intervals also includes: the current time interval takes the current time as the starting time; Among them, using the convolutional neural network model to intelligently predict a single fault microservice data corresponding to the overall system in the current time interval based on multiple copies of fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems also includes: synchronously inputting multiple copies of fault microservice data corresponding to the overall system in multiple past time intervals, deployment information of various microservices deployed by the overall system, the total number of various microservices deployed by the overall system, and the number of systems of the multiple information systems into the convolutional neural network model; running the convolutional neural network model to obtain a single copy of fault microservice data corresponding to the overall system in the current time interval output by the convolutional neural network model.

Images