Cloud computer monitoring method and device, equipment and storage medium
By acquiring the operational status information of cloud computers and using adaptive machine learning models to predict link quality and identify performance bottlenecks, the complexity of configuration and management for users in cloud computer use is solved, intelligent monitoring and resource optimization are achieved, and user experience and ecosystem development are improved.
Patent Information
- Application Number
- CN202410846457.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-22
AI Technical Summary
Users face complex configuration and management tasks when using cloud computers, which affects the user experience and hinders the development of the cloud computer ecosystem.
By acquiring the operating status information of cloud computers, we use adaptive machine learning models to predict link quality, identify performance bottlenecks, and generate alarm information to trigger response mechanisms for repair and resource optimization. At the same time, we recommend resources based on user behavior analysis.
It enables intelligent monitoring of cloud computer performance, improves operation and maintenance efficiency, reduces user configuration and management tasks, enhances user experience, and is conducive to the healthy development of the ecosystem.
Smart Images

Figure CN119003269A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of cloud computing technology, and in particular to a cloud computer monitoring method, device, equipment and storage medium. Background Art
[0002] As cloud computing technology becomes more mature and popularized, cloud computers, as a new computing model, are becoming more and more widely used due to their high efficiency, convenience and security. However, in the process of using and managing cloud computers, users usually need to face complex configuration and management tasks, which not only affects the user experience of cloud computers, but also restricts the further development of the cloud computer ecosystem. Summary of the invention
[0003] The present invention provides a cloud computer monitoring method, device, equipment and storage medium, which are used to solve the defect in the prior art that users need to face complex configuration and management tasks when using cloud computers, which affects the user's experience of using the cloud computer.
[0004] The present invention provides a cloud computer monitoring method, comprising: Obtaining the operating status information of the cloud computer; the operating status information includes at least one of hardware resource usage information, network connection information, and application program operation information; Determining the link quality of the cloud computer according to the operating status information; The performance bottleneck of the cloud computer is identified based on the link quality to monitor the performance of the cloud computer.
[0005] According to the cloud computer monitoring method provided by the present invention, determining the link quality of the cloud computer according to the operating status information includes: Extracting features from the running status information to obtain running features of the running status information; the running features include network performance indicators of the cloud computer; The operating characteristics are input into a pre-trained adaptive machine learning model, and the adaptive machine learning model is used to predict the link quality of the cloud computer; wherein the adaptive machine learning model is obtained by pre-training a preset basic machine learning model based on historical operating characteristics extracted from historical operating status information of the cloud computer and label data of link quality corresponding to the historical operating characteristics.
[0006] According to the cloud computer monitoring method provided by the present invention, before the operation characteristics are input into the pre-trained adaptive machine learning model and the link quality of the cloud computer is predicted by using the adaptive machine learning model, the method further includes: based on a preset training cycle, obtaining the historical operation status information of each cloud computer in the training cycle, and extracting the historical operation characteristics of each cloud computer from the historical operation status information; Generate label data of the link quality of the historical operation characteristics, and build a feature library based on the historical operation characteristics and the label data; The feature library is used to perform periodic iterative training on the preset basic machine learning model to obtain a pre-trained adaptive machine learning model.
[0007] According to the cloud computer monitoring method provided by the present invention, after identifying the performance bottleneck of the cloud computer based on the link quality, the method further includes: Generate alarm information according to the performance bottleneck, and send the alarm information to a preset alarm receiving terminal; A response mechanism to the performance bottleneck is triggered according to the alarm information to repair the performance bottleneck and optimize resources.
[0008] According to the cloud computer monitoring method provided by the present invention, after obtaining the operation status information of the cloud computer, the method further includes: performing behavior analysis on the target user corresponding to the cloud computer according to the operation status information to obtain the behavior data of the target user; Matching the behavior data with the resource library of the cloud computer to obtain a target resource corresponding to the behavior data; the target resource includes at least one of a cloud service, an application program, and a content resource; The target resource is recommended to the target user.
[0009] According to the cloud computer monitoring method provided by the present invention, the step of obtaining the operating status information of the cloud computer includes: Obtaining demand information of a target user corresponding to the cloud computer; the demand information is used to request to view at least one of the operating status, management configuration and resource library of the cloud computer; The operating status information of the cloud computer is obtained according to the demand information.
[0010] According to the cloud computer monitoring method provided by the present invention, after determining the link quality of the cloud computer according to the operating status information, the method further includes: Determining an optimal control strategy for the cloud computer according to the link quality; The cloud computer is allocated and adjusted in terms of resources according to the optimal control strategy.
[0011] The present invention also provides a cloud computer monitoring device, comprising the following modules: A data acquisition module, used to obtain the operation status information of the cloud computer; the operation status information includes at least one of hardware resource usage information, network connection information and application program operation information; A quality assessment module, used to determine the link quality of the cloud computer according to the operation status information; A performance monitoring module is used to identify the performance bottleneck of the cloud computer based on the link quality to monitor the performance of the cloud computer.
[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the cloud computer monitoring method as described in any one of the above are implemented.
[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the cloud computer monitoring methods described above.
[0014] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the cloud computer monitoring methods described above.
[0015] The cloud computer monitoring method, device, equipment and storage medium provided by the present invention determine the link quality of the cloud computer by acquiring the operating status information of the cloud computer, and identify the performance bottleneck of the cloud computer based on the link quality of the cloud computer, thereby realizing performance monitoring of the cloud computer. Based on the performance monitoring of the cloud computer, performance problems of the cloud computer can be discovered in a timely manner, the operation and maintenance and management efficiency of the cloud computer can be improved, the configuration and management tasks faced by users when using the cloud computer can be reduced, the user experience can be improved, and the healthy development of the cloud computer ecosystem can be facilitated. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0017] Figure 1 It is a flow chart of the cloud computer monitoring method provided by the present invention; Figure 2 It is a schematic diagram of the cloud computer monitoring process provided by the present invention; Figure 3It is a structural schematic diagram of the cloud computer monitoring device provided by the present invention; Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION
[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0019] In order to solve the relevant technical problems that users have when using cloud computers, an embodiment of the present invention provides an application program, which is embedded in the cloud computer as a "cloud computer management assistant" to realize the automatic and intelligent management of the cloud computer, and effectively promote the construction and development of the cloud computer ecosystem. The application program provides a cloud computer monitoring method, which uses intelligent operation and maintenance technology to conduct comprehensive and in-depth monitoring of the performance, security and reliability of the cloud computer.
[0020] Specifically, Figure 1 It is a flow chart of the cloud computer monitoring method provided by the present invention, such as Figure 1 As shown, the method comprises the following steps: Step 100, obtaining the operating status information of the cloud computer; the operating status information includes at least one of hardware resource usage information, network connection information and application program operating information; Step 200, determining the link quality of the cloud computer according to the operation status information; Step 300, identifying the performance bottleneck of the cloud computer based on the link quality to monitor the performance of the cloud computer.
[0021] Obtaining the operation status information of the cloud computer, the operation status information including at least one of hardware resource usage information, network connection information and application program operation information. The hardware resource usage information is used to characterize the hardware resource usage of the cloud computer, the network connection information is used to characterize the network connection of the cloud computer, and the application program operation information is used to characterize the operation of the application program in the cloud computer.
[0022] Optionally, the operating status information of the cloud computer is obtained based on the monitoring data of the cloud computer, and the monitoring data includes the full-link performance data, log information and business data of the cloud computer. By monitoring the cloud computer, the full-link performance data, log information and business data of the cloud computer are collected, and the operating status information of the cloud computer is determined based on the full-link performance data, log information and business data of the cloud computer.
[0023] Optionally, the cloud computer may include one or more cloud computers. When the cloud computer includes multiple cloud computers, each cloud computer is embedded with a "cloud computer management assistant" application, and the acquired operating status information includes the operating status information of multiple cloud computers.
[0024] The link quality of the cloud computer is determined based on the acquired operating status information. When there are multiple cloud computers, the link quality of any cloud computer is determined based on the operating status information of the cloud computer. Optionally, the link quality can be determined by performing statistical analysis on the operating status information to obtain the operating indicators of the cloud computer; or it can be determined by using a pre-trained state recognition model to identify the operating status information of the cloud computer to obtain the link quality of the cloud computer, which is not specifically limited here.
[0025] Based on the link quality of the cloud computer, the performance bottleneck of the cloud computer is identified, so as to realize the monitoring of the performance of the cloud computer. It can be seen that different link qualities correspond to different performance problems. By locating the root cause, the performance bottleneck of the cloud computer can be identified, so as to optimize it in a targeted manner.
[0026] Optionally, the recognition of the performance bottleneck of the cloud computer may also be implemented based on a pre-trained fault recognition model, and the fault recognition model and the state recognition model may be the same recognition model or different recognition models.
[0027] In one embodiment, the acquired operation status information of the cloud computer is input into a pre-trained monitoring model, which includes a feature extraction network, a link quality identification network, and a root cause location network. The feature extraction network is used to extract features from the operation status information, and the extracted operation features are input into the link quality identification network to identify the link quality of the cloud computer. The identified link quality is input into the root cause location network, and the root cause location network performs root cause location based on the link quality to identify the performance bottleneck of the cloud computer.
[0028] In this embodiment, the link quality of the cloud computer is determined by obtaining the operating status information of the cloud computer, and the performance bottleneck of the cloud computer is identified based on the link quality of the cloud computer to implement performance monitoring of the cloud computer. Based on the performance monitoring of the cloud computer, performance problems of the cloud computer can be discovered in a timely manner, and the operation and maintenance and management efficiency of the cloud computer can be improved. The configuration and management tasks faced by users when using the cloud computer can be reduced, the user experience can be improved, and it is conducive to the healthy development of the cloud computer ecosystem.
[0029] In one embodiment, the link quality of the cloud computer is identified based on a pre-trained adaptive machine learning model. Specifically, in step 200, the link quality of the cloud computer is determined according to the operating status information of the cloud computer, including: Step 210, extracting features from the operation status information to obtain operation features of the operation status information; the operation features include network performance indicators of the cloud computer; Step 220, input the operating characteristics into a pre-trained adaptive machine learning model, and use the adaptive machine learning model to predict the link quality of the cloud computer; wherein the adaptive machine learning model is obtained by pre-training a preset basic machine learning model based on historical operating characteristics extracted from historical operating status information of the cloud computer and label data of the link quality corresponding to the historical operating characteristics.
[0030] First, feature extraction is performed on the operating status information of the cloud computer to obtain the operating features of the operating status information, which include the network performance indicators of the cloud computer. The extracted operating features are input into a pre-trained adaptive machine learning model, and the adaptive machine learning model is used to predict the link quality of the cloud computer. The adaptive machine learning model is obtained by pre-training a preset basic machine learning model based on the historical operating features extracted from the historical operating status information of the cloud computer and the label data of the link quality corresponding to the historical operating features.
[0031] Optionally, the prediction of the link quality of the cloud computer includes the identification of the link quality of the cloud computer at the current moment, and also includes the prediction of the link quality of the cloud computer within a preset time period, which is a shorter time period.
[0032] Optionally, the network performance indicators of cloud computers include boot time, network latency, and resource utilization.
[0033] In one embodiment, the training of the adaptive machine learning model is a periodic automatic process. In step 220, before the extracted operation features are input into the pre-trained adaptive machine learning model, the following steps may also be included: Step 2101, based on a preset training cycle, obtaining historical operation status information of each cloud computer in the training cycle, and extracting historical operation features of each cloud computer from the historical operation status information; Step 2102, generating label data of the link quality of the historical operation characteristics, and building a feature library based on the historical operation characteristics and the label data; Step 2103, use the feature library to periodically iterate the preset basic machine learning model to obtain a pre-trained adaptive machine learning model.
[0034] In the model training phase, based on the preset training cycle, the historical operation status information of each cloud computer in the training cycle is obtained, and the historical operation characteristics of the cloud computer are extracted from the obtained historical operation status information. According to the annotation of the cloud computer operation characteristics, the label data of the link quality of the historical operation characteristics is generated, and a feature library is built based on the historical operation characteristics and their label data. The preset basic machine learning model is periodically iterated using the feature library to obtain a pre-trained adaptive machine learning model.
[0035] That is, based on the preset training cycle, after a certain period of time, the preset basic machine learning model is iteratively trained based on the latest operating status information of the cloud computer. Optionally, the model training of any training cycle is performed on the basis of the adaptive machine learning model obtained by training in the previous training cycle.
[0036] Optionally, the constructed feature library may include only the historical operation features and their label data obtained in the most recent training cycle, or may include the historical operation features and their label data obtained in each training cycle. Periodic training of the adaptive machine learning model helps the model to adapt to the performance changes of cloud computers and user behavior changes in a timely manner, continuously learn and adapt to changes in the business environment, ensure the effectiveness and accuracy of the model recognition results, and continuously improve the accuracy and efficiency of monitoring and alarming.
[0037] In one embodiment, for the performance bottleneck of the cloud computer, a corresponding response mechanism can be triggered to achieve self-repair of the performance bottleneck. In step 300, after identifying the performance bottleneck of the cloud computer based on the link quality, the following steps can also be included: Step 301, generate alarm information according to the performance bottleneck, and send the alarm information to a preset alarm receiving terminal; Step 302, trigger a response mechanism to the performance bottleneck according to the alarm information to repair the performance bottleneck and optimize resources.
[0038] An alarm message is generated according to the performance bottleneck, and the generated alarm message is sent to a preset alarm receiving terminal. At the same time, a response mechanism to the performance bottleneck is triggered according to the generated alarm message to achieve automatic repair and resource optimization of the identified performance bottleneck.
[0039] Optional self-repair and resource optimization of cloud computer performance bottlenecks, including but not limited to automatic cleaning of memory and junk files, and repair of detected high-risk vulnerabilities when a performance bottleneck of the cloud computer is detected; or, the user can trigger the cleaning of memory and junk files, and repair of high-risk vulnerabilities with one click according to the repair prompts in the alarm information, thereby realizing convenient self-operation and maintenance of the cloud computer.
[0040] In one embodiment, after determining the link quality of the cloud computer, resource optimization may be performed according to the link quality of the cloud computer. After step 200, the following may also be included: Step 201, determining the optimal control strategy of the cloud computer according to the link quality; Step 202: Allocate and adjust resources of the cloud computer according to the optimal control strategy.
[0041] The optimal control strategy of the cloud computer is determined according to the link quality of the cloud computer. Optionally, the optimal control strategy of the cloud computer may include a resource adjustment strategy and a load strategy. According to the optimal control strategy, resources of the cloud computer are allocated and adjusted, resource configuration and operation strategy are intelligently adjusted, and efficient and stable operation of the cloud computer is ensured.
[0042] In some embodiments, after obtaining the operating status information of the cloud computer, the user behavior can be analyzed according to the operating status information of the cloud computer, so as to achieve accurate recommendations for the user. After step 100, the following may also be included: Step 101, performing behavior analysis on a target user corresponding to the cloud computer according to the operation status information to obtain behavior data of the target user; Step 102, matching the behavior data with the resource library of the cloud computer to obtain a target resource corresponding to the behavior data; the target resource includes at least one of a cloud service, an application program, and a content resource; Step 103: recommend the target resource to the target user.
[0043] Based on the acquired operating status information of the cloud computer, a behavior analysis is performed on the target user corresponding to the cloud computer to obtain the behavior data of the target user, and the behavior data of the target user is matched with the resource library of the cloud computer to obtain the target resource corresponding to the behavior data, which includes at least one of cloud services, applications and content resources, and the matched target resource is recommended to the target user.
[0044] Optionally, based on the operating status information of the cloud computer, the applications, cloud services, and content resources frequently used by the cloud computer users can be determined. The user's behavior data is data used to characterize the user's use behavior of the cloud computer, including but not limited to the user's use frequency and use duration of at least one of the applications, cloud services, and content resources. Based on the user's behavior data of using the cloud computer, target resources that match the user's behavior data are matched from the cloud computer's resource library and recommended to the user.
[0045] Optionally, behavior analysis of target users corresponding to the cloud computer, including behavior event analysis, page click analysis, behavior path analysis, user health analysis and user portrait analysis.
[0046] Among them, behavior event analysis is used to analyze the impact of specific user behavior events on cloud computers and the extent of the impact. Events can be obtained through tracking points, including the definition of the participants, time, location, interaction method and content of the event. For example, for the login page, the difference in usage behavior between new users who click to log in and those who skip login can be analyzed, so as to optimize the login process and improve the user experience.
[0047] Page click analysis is used to analyze the user's click position and click frequency on the page, find out the content that the user is interested in on the current page, evaluate the rationality of the page layout, and optimize the presentation of key content.
[0048] Behavioral path analysis tracks the user's specific behavioral path in the product, understands the user's usage habits and preferences, and finds the key points to optimize the user experience.
[0049] User health analysis evaluates the overall health of the user group and identifies problems in a timely manner by monitoring indicators such as user activity and retention rate. Indicators that need to be monitored include daily active users (DAU), monthly active users (MAU), and the ratio of DAU to MAU, which can evaluate the overall activity and stickiness of users.
[0050] User portrait analysis collects and analyzes users’ basic information and behavior data to build user portraits, understand users’ needs and preferences, and provide personalized services. For example, by analyzing users’ browsing history, purchase records and other data, we can recommend products or services that users may be interested in, thereby improving user satisfaction and consumption frequency.
[0051] In this embodiment, there are many types of user behaviors for using the cloud computer, which can be specifically divided according to different usage scenarios of the users, where the usage scenarios mainly include office, entertainment, professional production, scientific research and education, etc. In different usage scenarios, different applications are used, so the user's use behavior of the cloud computer can include the use records of each application in the cloud computer, the number of uses, the duration of use, and the application type to which the used application belongs, etc., and can further include the user's browsing records of resources in the cloud computer, the browsing duration, and the number of clicks, etc.
[0052] Furthermore, based on the operating status information of the cloud computer, the behavior analysis of the target users of the cloud computer can be conducted. On the one hand, the user's usage habits of the cloud computer can be analyzed, and on the other hand, the user's potential demand for the cloud computer can be analyzed.
[0053] Optionally, the resource library of the cloud computer is obtained by integrating the resource pools of various resources of the cloud computer. Through intelligent matching and resource recommendation, corresponding resources can be accurately recommended to users based on their potential needs, thereby achieving the purpose of precision marketing and improving user experience and satisfaction.
[0054] In some embodiments, the target resources recommended to the user may be of different resource types, for example, the target resource may be news information, or a product resource of a cloud computer, or an application that the user may be interested in. In the recommendation information for the target resource, a corresponding interface may be provided, and the user may directly enter the ordering page of the target resource by clicking on the link corresponding to the interface.
[0055] For example, if the target resource is an application, the user can directly enter the cloud computer's application store to install and use the application after clicking the link corresponding to the interface; if the target resource is news information, the user can enter the news information details page to browse the details after clicking the connection corresponding to the interface; if the target resource is a cloud computer product, the user can enter the cloud computer's resource ordering page to order the product after clicking the connection corresponding to the interface.
[0056] Optionally, this interface can also be used to connect cloud computers to third-party platforms, thereby achieving cross-platform interoperability of cloud computers, improving user traffic conversion and sharing efficiency, and improving the cloud computer ecosystem.
[0057] Optionally, the behavior analysis of the target users corresponding to the cloud computer can be implemented based on a preset behavior recognition model. The behavior recognition model can be obtained by pre-training a preset basic deep learning model based on the historical operating status information of the cloud computer and the behavior labels of the historical operating status information as sample data.
[0058] In some embodiments, the operating status information of the cloud computer is obtained according to user needs, so that the cloud computer can be monitored according to the user's monitoring needs. Based on this, in step 100, the operating status information of the cloud computer is obtained, which specifically includes: Step 110, obtaining demand information of a target user corresponding to the cloud computer; the demand information is used to request to view at least one of the operating status, management configuration and resource library of the cloud computer; Step 120: Obtain the operating status information of the cloud computer according to the demand information.
[0059] Obtain demand information of a target user corresponding to the cloud computer, where the demand information is used to request to view at least one of the operating status, management configuration, and resource library of the cloud computer, and obtain the operating status information of the cloud computer according to the user's demand information.
[0060] Optionally, the acquired operating status information is status information associated with the target user's demand information, that is, status information associated with at least one of the operating status, management configuration, and resource library of the cloud computer requested by the target user to view.
[0061] Optionally, to facilitate the use of cloud computers by users, the cloud computer can display information such as the boot time, network latency, resource utilization, and renewal reminders of the cloud computer in real time according to the user's demand information. At the same time, according to the user's potential needs, it can also increase the recommendation of news information, applications, and cloud computer resources or products to achieve business reach of marketing services.
[0062] In one embodiment, referring to Figure 2 The cloud computer monitoring process shown in the figure includes several main stages: data collection, data preprocessing, feature extraction, model training and optimization, link quality assessment and prediction, alarm generation and intelligent response, and visual monitoring.
[0063] Among them, in the data collection stage, it is mainly used to collect the full-link performance data, log information and business data of the cloud computer to obtain the operating status information such as the hardware resource usage information, network connection information and application running information of the cloud computer. The operating status information of the cloud computer collected in the data collection stage includes real-time data and historical data. Optionally, the real-time data is used to monitor the performance of the cloud computer, and the real-time data and historical data are used to train the monitoring model of the cloud computer, that is, the adaptive machine learning model. Among them, the historical data of the operating status information of the cloud computer can be the real-time data of the operating status information of the cloud computer collected during the historical monitoring process of the cloud computer.
[0064] In the data preprocessing stage, the data collected in the data collection stage is preprocessed, which includes data cleaning, filtering, denoising and standardization. Data preprocessing is conducive to improving data quality and availability. The data collected in the data collection stage is preprocessed, including the preprocessing of real-time data of the cloud computer's operating status information and the preprocessing of historical data of the cloud computer's operating status.
[0065] Furthermore, in the model training and optimization phase, a feature library is constructed based on the operating status information of the cloud computer collected in the data collection phase, and the feature library is used to periodically train the adaptive machine learning model. The model training is carried out by combining the historical data and real-time data of the operating status information of the cloud computer to achieve adaptive learning and optimization of the model and continuously optimize the model performance. It can be seen that the training of the adaptive machine learning model is a periodic online training process. Through online learning and incremental learning, the adaptive machine learning model can continuously learn and adapt to changes in the business environment and user behavior, and continuously improve the accuracy of monitoring and alarming of cloud computers.
[0066] In the feature extraction stage, feature extraction is performed on the real-time data of the cloud computer's operating status information collected in the data collection stage. Technologies such as big data analysis and feature engineering are used to extract key features and indicators in the operating status information to obtain the operating characteristics of the cloud computer.
[0067] In the link quality evaluation and prediction stage, the extracted cloud computer operation characteristics are input into the pre-trained adaptive machine learning model, and the pre-trained adaptive machine learning model is used to evaluate and predict the full link quality of the cloud computer to obtain the link quality of the cloud computer. At the same time, the performance bottleneck of the cloud computer is identified based on the predicted link quality.
[0068] In the alarm generation and response stage, alarm information is generated based on the performance bottleneck of the cloud computer identified, and a detailed alarm is issued for the performance bottleneck of the cloud computer. At the same time, the alarm information is sent to the preset alarm receiving terminal through multiple channels. Furthermore, the intelligent response mechanism for the performance bottleneck is triggered based on the generated alarm information, realizing automatic repair of the performance bottleneck and resource optimization.
[0069] In the visual monitoring stage, a visual monitoring interface is provided to visualize the real-time status, historical trends, and alarm information of the link quality of the entire link of the cloud computer on the monitoring interface, which is convenient for operation and maintenance personnel to monitor and manage. Among them, the historical trend of link quality can be obtained based on the historical monitoring process of the performance of the cloud computer. The alarm information generated according to the performance bottleneck includes the alarm type and alarm level. The alarm type is used to characterize the fault type of the performance bottleneck of the cloud computer, and the alarm level is used to characterize the fault degree of the performance bottleneck of the cloud computer.
[0070] In some embodiments, user feedback information can be obtained in the same manner as user demand information, and the performance of the cloud computer can be continuously improved and optimized based on the user feedback information, thereby accurately improving user satisfaction.
[0071] In this embodiment, through periodic online training of the adaptive machine learning model, the adaptive machine learning model can timely adapt to the performance changes of the cloud computer and the behavior changes of the user during the performance monitoring of the cloud computer, and continuously improve the accuracy of monitoring and alarming of the cloud computer through continuous learning.
[0072] Furthermore, through analysis of user behavior, we can accurately recommend resources that meet users' needs based on their preferences and potential needs, thereby helping to improve users' experience and satisfaction with cloud computers.
[0073] Furthermore, based on the performance monitoring of cloud computers, the resource configuration of cloud computers can be adjusted in real time according to the operating status of cloud computers. On the one hand, it can ensure the efficient and stable operation of cloud computers, and on the other hand, it can improve the management efficiency of cloud computers, reduce the need for manual intervention, and reduce labor costs and risks caused by human errors.
[0074] The cloud computer monitoring device provided by the present invention is described below. The cloud computer monitoring device described below and the cloud computer monitoring method described above can be referenced to each other.
[0075] Reference Figure 3 , the cloud computer monitoring device provided by the embodiment of the present invention includes: The data acquisition module 10 is used to obtain the operation status information of the cloud computer; the operation status information includes at least one of hardware resource usage information, network connection information and application program operation information; A quality assessment module 20, configured to determine the link quality of the cloud computer according to the operation status information; The performance monitoring module 30 is used to identify the performance bottleneck of the cloud computer based on the link quality so as to monitor the performance of the cloud computer.
[0076] In one embodiment, the data acquisition module 10 is further used for: Obtaining demand information of a target user corresponding to the cloud computer; the demand information is used to request to view at least one of the operating status, management configuration and resource library of the cloud computer; The operating status information of the cloud computer is obtained according to the demand information.
[0077] In one embodiment, the quality assessment module 20 is further configured to: Extracting features from the running status information to obtain running features of the running status information; the running features include network performance indicators of the cloud computer; The operating characteristics are input into a pre-trained adaptive machine learning model, and the adaptive machine learning model is used to predict the link quality of the cloud computer; wherein the adaptive machine learning model is obtained by pre-training a preset basic machine learning model based on historical operating characteristics extracted from historical operating status information of the cloud computer and label data of link quality corresponding to the historical operating characteristics.
[0078] In one embodiment, the cloud computer monitoring device further includes a model training module for: Based on a preset training cycle, historical operation status information of each cloud computer in the training cycle is obtained, and historical operation characteristics of each cloud computer are extracted from the historical operation status information; Generate label data of the link quality of the historical operation characteristics, and build a feature library based on the historical operation characteristics and the label data; The feature library is used to perform periodic iterative training on the preset basic machine learning model to obtain a pre-trained adaptive machine learning model.
[0079] In one embodiment, the cloud computer monitoring device further includes an alarm response module for: Generate alarm information according to the performance bottleneck, and send the alarm information to a preset alarm receiving terminal; A response mechanism to the performance bottleneck is triggered according to the alarm information to repair the performance bottleneck and optimize resources.
[0080] In one embodiment, the cloud computer monitoring device further includes a resource recommendation module for: Performing behavior analysis on a target user corresponding to the cloud computer according to the operation status information to obtain behavior data of the target user; Matching the behavior data with the resource library of the cloud computer to obtain a target resource corresponding to the behavior data; the target resource includes at least one of a cloud service, an application program, and a content resource; The target resource is recommended to the target user.
[0081] In one embodiment, the cloud computer monitoring device further includes a resource tuning module for: Determining an optimal control strategy for the cloud computer according to the link quality; The cloud computer is allocated and adjusted in terms of resources according to the optimal control strategy.
[0082] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other via the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the steps of the cloud computer monitoring method, for example including: Obtaining the operating status information of the cloud computer; the operating status information includes at least one of hardware resource usage information, network connection information, and application program operation information; Determining the link quality of the cloud computer according to the operating status information; The performance bottleneck of the cloud computer is identified based on the link quality to monitor the performance of the cloud computer.
[0083] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0084] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, when the computer program is executed by a processor, the computer can perform the steps of the cloud computer monitoring method provided by the above methods, for example, including: Obtaining the operating status information of the cloud computer; the operating status information includes at least one of hardware resource usage information, network connection information, and application program operation information; Determining the link quality of the cloud computer according to the operating status information; The performance bottleneck of the cloud computer is identified based on the link quality to monitor the performance of the cloud computer.
[0085] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the cloud computer monitoring method provided by the above methods, for example, including: Obtaining the operating status information of the cloud computer; the operating status information includes at least one of hardware resource usage information, network connection information, and application program operation information; Determining the link quality of the cloud computer according to the operating status information; The performance bottleneck of the cloud computer is identified based on the link quality to monitor the performance of the cloud computer.
[0086] 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 displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0087] 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 necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product 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 embodiments.
[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 invention.
Claims
1. A cloud computer monitoring method, characterized in that: include: Get the running status information of the cloud computer; The operation status information includes at least one of hardware resource usage information, network connection information and application program operation information; Determining the link quality of the cloud computer according to the operating status information; The performance bottleneck of the cloud computer is identified based on the link quality to monitor the performance of the cloud computer.
2. The cloud computer monitoring method according to claim 1, characterized in that: Determining the link quality of the cloud computer according to the operating status information includes: Extracting features from the running status information to obtain running features of the running status information; the running features include network performance indicators of the cloud computer; The operating characteristics are input into a pre-trained adaptive machine learning model, and the adaptive machine learning model is used to predict the link quality of the cloud computer; wherein the adaptive machine learning model is obtained by pre-training a preset basic machine learning model based on historical operating characteristics extracted from historical operating status information of the cloud computer and label data of link quality corresponding to the historical operating characteristics.
3. The cloud computer monitoring method according to claim 2, characterized in that: Before inputting the operation characteristics into a pre-trained adaptive machine learning model and using the adaptive machine learning model to predict the link quality of the cloud computer, the method further includes: Based on a preset training cycle, historical operation status information of each cloud computer in the training cycle is obtained, and historical operation characteristics of each cloud computer are extracted from the historical operation status information; Generate label data of the link quality of the historical operation characteristics, and build a feature library based on the historical operation characteristics and the label data; The feature library is used to perform periodic iterative training on the preset basic machine learning model to obtain a pre-trained adaptive machine learning model.
4. The cloud computer monitoring method according to claim 1, characterized in that: After identifying the performance bottleneck of the cloud computer based on the link quality, the method further includes: Generate alarm information according to the performance bottleneck, and send the alarm information to a preset alarm receiving terminal; A response mechanism to the performance bottleneck is triggered according to the alarm information to repair the performance bottleneck and optimize resources.
5. The cloud computer monitoring method according to claim 1, characterized in that: After obtaining the operating status information of the cloud computer, the method further includes: Performing behavior analysis on a target user corresponding to the cloud computer according to the operation status information to obtain behavior data of the target user; Matching the behavior data with the resource library of the cloud computer to obtain a target resource corresponding to the behavior data; the target resource includes at least one of a cloud service, an application program, and a content resource; The target resource is recommended to the target user.
6. The cloud computer monitoring method according to claim 1, characterized in that: The obtaining of the operating status information of the cloud computer includes: Obtaining demand information of a target user corresponding to the cloud computer; the demand information is used to request to view at least one of the operating status, management configuration and resource library of the cloud computer; The operating status information of the cloud computer is obtained according to the demand information.
7. The cloud computer monitoring method according to claim 1, characterized in that: After determining the link quality of the cloud computer according to the operating status information, the method further includes: Determining an optimal control strategy for the cloud computer according to the link quality; The cloud computer is allocated and adjusted in terms of resources according to the optimal control strategy.
8. A cloud computer monitoring device, characterized in that: include: Data collection module, used to obtain the operating status information of the cloud computer; The operation status information includes at least one of hardware resource usage information, network connection information and application program operation information; A quality assessment module, used to determine the link quality of the cloud computer according to the operation status information; A performance monitoring module is used to identify the performance bottleneck of the cloud computer based on the link quality to monitor the performance of the cloud computer.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the cloud computer monitoring method as described in any one of claims 1 to 7 are implemented.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the cloud computer monitoring method as described in any one of claims 1 to 7 are implemented.
