Intelligent operation and maintenance method and system for server

Abstract

The invention relates to the technical field of server operation and maintenance, in particular to an intelligent server operation and maintenance method and system. The method comprises the following steps: acquiring server hardware state data, extracting index characteristics, setting a threshold range for each index, and constructing a monitoring index system; acquiring hardware state data of the server in real time, comparing the real-time data with the monitoring index system, and evaluating the current running state and fault risk of the server; an operation and maintenance strategy library is set, an operation and maintenance decision is generated and executed according to the diagnosis data, and the execution progress and result are obtained in real time; the system comprises a monitoring index system construction module, a fault prediction and diagnosis module and an operation and maintenance decision execution feedback module. According to the method, the monitoring index system is constructed according to the index characteristics, and the data machine evaluates and generates the corresponding operation and maintenance decision according to the real-time hardware state of the server, so that the operation state and the potential fault risk of the server can be identified more accurately, the subjectivity and the error of manual judgment are avoided, and the continuity and the stability of operation and maintenance are improved.

Description

Technical Field

[0001] This invention relates to the field of server operation and maintenance technology, and in particular to a server intelligent operation and maintenance method and system. Background Technology

[0002] With the rapid development of information technology, servers, as core devices for data storage, processing, and transmission, play a crucial role in the informatization of various fields. Their stable and reliable operation directly affects the continuity of business operations, data security, and service quality. However, during long-term operation, server hardware is inevitably affected by various factors, such as ambient temperature, humidity, voltage fluctuations, and equipment aging, leading to hardware failures. Once server hardware fails, it can cause serious consequences such as data loss and service interruption, resulting in significant economic losses and reputational damage to enterprises.

[0003] Currently, traditional operations and maintenance mainly rely on the experience of operations and maintenance personnel and manual operation, which is inefficient, prone to human error, and difficult to monitor the server's various performance indicators and operating status in real time. It is also difficult to detect potential problems in a timely manner, and repairs can only be carried out after a failure occurs, which affects the continuity and stability of operations and maintenance. Summary of the Invention

[0004] The purpose of this invention is to provide a server intelligent operation and maintenance method and system, which aims to solve the technical problems of low efficiency, easy human error, difficulty in real-time monitoring of various performance indicators and operating status of servers, inability to detect potential problems in a timely manner, and often only being able to repair after a failure occurs, thus affecting the continuity and stability of operation and maintenance.

[0005] To achieve the above objectives, the present invention employs a server intelligent operation and maintenance method, comprising the following steps:

[0006] Acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system;

[0007] Deploy a data acquisition agent to collect server hardware status data in real time, compare real-time data with monitoring indicator system, assess the current operating status and failure risk of the server, and output diagnostic data;

[0008] Set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

[0009] Among the steps involved in acquiring server hardware status data, extracting indicator features, setting threshold ranges for each indicator, and constructing a monitoring indicator system:

[0010] The server's management interface is used to collect multi-dimensional status data of the server hardware and preprocess the multi-dimensional status data.

[0011] Extract static, dynamic, and correlation indicator features from multi-dimensional state data, and filter out redundant data in the indicator features;

[0012] Each indicator characteristic has a set range of basic threshold, early warning threshold and fault threshold, and the thresholds are dynamically adjusted based on historical data;

[0013] By integrating indicator data, threshold ranges, and data sources, a monitoring indicator system is constructed.

[0014] Among the steps, the following steps involve collecting multi-dimensional status data of the server hardware through the server's management interface and preprocessing this multi-dimensional status data:

[0015] Clean the data, remove noise and outliers, identify missing data, and complete the data accordingly.

[0016] Set a collection period. When there is a short-term data gap, use the average of data from adjacent periods to complete the data.

[0017] When data is missing for a long period of time, cache the file and fill in the historical data.

[0018] In the step of extracting static, dynamic, and correlated indicator features from multi-dimensional state data and filtering out redundant data in the indicator features, the process of extracting correlated indicator features is as follows:

[0019] Establish an indicator correlation model to analyze the interrelationships between multiple static and dynamic indicators, and create an indicator combination, which serves as the feature of the correlation indicator.

[0020] Among the steps involved are: deploying a data acquisition agent to collect server hardware status data in real time, comparing real-time data with monitoring indicator systems, assessing the current operating status and failure risks of the server, and outputting diagnostic data;

[0021] Deploy the data acquisition agent and configure its settings, including the data acquisition frequency, list of acquisition metrics, and data transmission target address.

[0022] Continuously collect server hardware status data based on the data collection frequency;

[0023] Extract real-time indicator data and compare the real-time indicator data with the corresponding threshold range in the monitoring indicator system;

[0024] Based on the comparison results of various indicators, the overall operating status of the server hardware is evaluated, and comprehensive status data is output.

[0025] Obtain the trend of indicator changes, predict the server failure risk based on historical data and indicator change trends, and output failure risk data.

[0026] After the step of continuously collecting server hardware status data according to the data collection frequency:

[0027] A virtual buffer is established to receive server hardware status data, assign timestamps to the data, preprocess the received data, and store the data based on the timestamps.

[0028] The process includes the steps of acquiring indicator change trends, predicting server failure risks based on historical data and indicator change trends, and outputting failure risk data:

[0029] Based on comprehensive status data and fault risk data, a current server diagnostic report is generated and visualized.

[0030] Among the steps are: setting up an operation and maintenance strategy library, generating and executing operation and maintenance decisions based on diagnostic data, obtaining execution progress and results in real time, and receiving feedback information;

[0031] Define strategy categories and configure strategy parameters, set up an operation and maintenance strategy library, and optimize the operation and maintenance strategy library;

[0032] Establish a decision generation engine to generate operation and maintenance decisions based on diagnostic data, and execute the decisions according to the process;

[0033] Establish a feedback mechanism to collect information and feedback from operations and maintenance personnel during the implementation of operations and maintenance decisions.

[0034] This includes the steps of establishing a decision generation engine, generating operational decisions based on diagnostic data, and executing those decisions according to the established process:

[0035] Get execution progress information in real time and verify the execution results.

[0036] This invention also provides a server intelligent operation and maintenance system, including a monitoring indicator system construction module, a fault prediction and diagnosis module, and an operation and maintenance decision execution feedback module; wherein:

[0037] The monitoring indicator system construction module is used to acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system.

[0038] The fault prediction and diagnosis module is used to deploy a data acquisition agent to collect server hardware status data in real time, compare real-time data with the monitoring indicator system, assess the current operating status and fault risk of the server, and output diagnostic data.

[0039] The operation and maintenance decision execution feedback module is used to set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

[0040] This invention discloses a server intelligent operation and maintenance method and system, which employs a monitoring indicator system construction module, a fault prediction and diagnosis module, and an operation and maintenance decision execution feedback module to perform the following steps: acquiring server hardware status data, extracting indicator features, setting threshold ranges for each indicator, and constructing a monitoring indicator system; deploying a data acquisition agent to collect server hardware status data in real time, comparing real-time data with the monitoring indicator system, assessing the current operating status and fault risks of the server, and outputting diagnostic data; setting an operation and maintenance strategy library, generating and executing operation and maintenance decisions based on the diagnostic data, acquiring execution progress and results in real time, and receiving feedback information; by constructing a monitoring indicator system based on indicator features and evaluating it based on real-time server hardware status data to generate corresponding operation and maintenance decisions, the system can more accurately identify the server's operating status and potential fault risks, avoiding the subjectivity and errors of manual judgment, and improving the continuity and stability of operation and maintenance. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart of the steps of the intelligent server operation and maintenance method of the present invention.

[0043] Figure 2 This is a flowchart of steps S100 of the present invention.

[0044] Figure 3 This is a flowchart of steps S200 of the present invention.

[0045] Figure 4 This is a flowchart of steps S300 of the present invention.

[0046] Figure 5 This is a structural schematic diagram of the server intelligent operation and maintenance system of the present invention.

[0047] Figure 6 This is a schematic diagram of the electronic device of the present invention.

[0048] 401 - Monitoring indicator system construction module, 402 - Fault prediction and diagnosis module, 403 - Operation and maintenance decision execution feedback module. Detailed Implementation

[0049] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0050] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0051] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0052] Please see Figures 1-4 This invention provides a server intelligent operation and maintenance method, comprising the following steps:

[0053] S100: Acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system.

[0054] In this embodiment, server hardware status data is acquired, indicator features are extracted, threshold ranges are set for each indicator, and a monitoring indicator system is constructed. The specific process is as follows:

[0055] S101: Collect multi-dimensional status data of server hardware through the server's management interface, and preprocess the multi-dimensional status data.

[0056] S102: Extract static, dynamic, and correlation indicator features from multi-dimensional state data, and filter out redundant data in the indicator features;

[0057] S103: Set the range of basic threshold, early warning threshold and fault threshold for each indicator feature, and dynamically adjust the thresholds based on historical data;

[0058] S104: Integrate indicator data, threshold ranges, and data sources to construct a monitoring indicator system.

[0059] Furthermore, in the steps of collecting multi-dimensional status data of the server hardware through the server's management interface and preprocessing the multi-dimensional status data:

[0060] Clean the data, remove noise and outliers, identify missing data, and complete the data accordingly.

[0061] Set a collection period. When there is a short-term data gap, use the average of data from adjacent periods to complete the data.

[0062] When data is missing for a long period of time, cache the file and fill in the historical data.

[0063] Furthermore, in the steps of extracting static, dynamic, and correlated indicator features from multi-dimensional state data and filtering out redundant data in the indicator features, the process of extracting correlated indicator features is as follows:

[0064] Establish an indicator correlation model to analyze the interrelationships between multiple static and dynamic indicators, and create an indicator combination, which serves as the feature of the correlation indicator.

[0065] In the above process, data is sourced through server management interfaces such as IPMI (Intelligent Platform Management Interface) and BMC (Baseboard Management Controller), which provide direct access to the server hardware status. Alternatively, tools and commands provided by the operating system can be utilized. For example, in Linux systems, the `lshw` command can obtain detailed hardware information, the `dmidecode` command can read hardware data from the DMI (Desktop Management Interface) table, including information on the motherboard, memory, and CPU, and the `sensors` command can obtain hardware sensor data such as temperature and voltage.

[0066] Server hardware status data is collected from multiple dimensions, covering CPU utilization, temperature, and frequency; memory usage and error rate; disk read / write speed, remaining space, and I / O errors; network card traffic and packet loss rate; and key indicators such as power supply voltage, current, and power. This ensures that the collected data comprehensively reflects the operating status of the server hardware.

[0067] Extracting key features and preprocessing the data: Cleaning the collected raw data to remove noise and outliers. For example, temperature data collected by sensors may contain some unreasonable high or low values ​​due to external interference or sensor malfunction. These values ​​can be filtered and corrected by setting reasonable ranges.

[0068] Feature extraction methods: Statistical analysis is used to calculate the average, maximum, minimum, and standard deviation of each indicator. These features reflect the central tendency, dispersion, and fluctuation of the indicator. For example, the average CPU utilization reflects the overall computing load of the server, while the standard deviation reflects the magnitude of fluctuations in utilization. Furthermore, for some periodically changing indicators, such as disk read / write operations, frequency characteristics can be extracted using methods like Fourier transform to better understand their patterns of change.

[0069] Basic threshold setting: Based on official documents such as technical manuals and product specifications provided by server hardware manufacturers, obtain the theoretical parameter range of each hardware indicator under normal operating conditions. For example, for CPU temperature, the manufacturer may clearly state that its normal operating temperature should be between 30℃ and 60℃ under normal load. This range can then be used as the basic threshold range for CPU temperature indicators.

[0070] Warning threshold setting:

[0071] Based on the proportional adjustment of the base threshold, the warning threshold is set as a certain percentage upper limit of the base threshold. For example, for CPU utilization, if the upper limit of the base threshold is 80%, then the warning threshold can be set to 85% to 90%. When the CPU utilization reaches this range, the system issues a warning signal to remind maintenance personnel to pay attention to the server load.

[0072] Considering safety margins and taking into account the performance characteristics of server hardware and business needs, a certain safety margin is set as a warning threshold for each indicator. For example, for server power supply, considering the potential increase in load due to future business growth, a margin of 10% to 15% is reserved on top of the basic threshold as a warning threshold. When the power supply approaches this margin range, a timely warning is issued to avoid server failure due to insufficient power.

[0073] Historical anomaly data analysis involves examining minor anomalies that occurred during server hardware operation to identify the numerical ranges of indicators that caused these anomalies, which can then be used as a reference for warning thresholds. For example, if historical data analysis reveals that network performance begins to fluctuate slightly when server network bandwidth utilization reaches 75%, then 75% can be used as a warning threshold for network bandwidth utilization.

[0074] Fault threshold setting:

[0075] Hardware limit parameters refer to the technical limits of server hardware, using the values ​​of indicators that approach or reach the physical limits of the hardware as fault thresholds. For example, for CPU temperature, if its maximum safe operating temperature is 95℃, then when the CPU temperature reaches 95℃, the system determines it to be in a fault state and immediately takes corresponding protective measures, such as frequency reduction or shutdown.

[0076] Critical fault correlation analysis examines the performance of various indicators in server hardware during severe faults, identifying the range of indicator values ​​closely related to the severe fault as fault thresholds. For example, when a server hard drive experiences severe faults such as an increase in bad sectors and frequent read / write errors, it is usually accompanied by a sharp increase in certain parameters of the hard drive's SMART indicator (such as the number of sectors to be reallocated and the number of sectors to be mapped). The critical values ​​of these parameters are used as fault thresholds; once these values ​​are reached, the hard drive is considered to have failed.

[0077] Thresholds are dynamically adjusted based on historical data, and a data collection mechanism is established: Historical data for each indicator is continuously collected through a server hardware monitoring system, including indicator values, collection time, server load, and other information. The data collection frequency is ensured to be sufficiently high to accurately reflect the changing trends of the indicators. For example, for key indicators such as CPU utilization and memory utilization, data can be collected every minute.

[0078] Data cleaning and preprocessing: The collected historical data is cleaned to remove noise and outliers. For example, data loss or errors due to monitoring system malfunctions or network outages require appropriate processing, such as filling in missing values ​​and correcting outliers. Simultaneously, the data is normalized to convert indicators with different dimensions to a uniform range, facilitating subsequent analysis and comparison.

[0079] Data analysis and model building involve using time series analysis to analyze the historical data of various indicators and identify patterns of change over time. For example, by plotting a line graph of CPU utilization, we can observe its trend across different time periods (such as daytime, nighttime, weekdays, and weekends) to determine whether there are cyclical changes or long-term growth / decline trends.

[0080] Correlation analysis examines the relationships between various indicators to identify related combinations. For example, CPU utilization and memory utilization are often correlated; when CPU utilization increases, memory utilization may also increase. Correlation analysis provides a more comprehensive understanding of the server hardware's operational status, offering a more accurate basis for threshold adjustments.

[0081] Develop predictive models based on historical data and employ machine learning algorithms (such as regression analysis and neural networks) to predict the changes in the values ​​of various indicators over a future period. For example, use historical CPU usage data to build a regression model to predict CPU usage within the next hour, and adjust the thresholds in advance based on the prediction results.

[0082] The threshold dynamic adjustment strategy is based on trend analysis: the thresholds are dynamically adjusted according to the trend analysis results of the indicators. If the indicator shows a long-term upward trend and the growth rate is relatively fast, the warning threshold and fault threshold can be appropriately increased to avoid frequent false alarms; conversely, if the indicator shows a downward trend, the threshold can be appropriately decreased to detect potential problems more promptly.

[0083] Adjustments based on predictions: Thresholds are adjusted in advance based on the prediction results of the indicator prediction model. For example, if it is predicted that CPU utilization will rise sharply in the next hour, approaching the warning threshold, the warning threshold can be appropriately raised in advance. At the same time, the changes in CPU utilization are closely monitored to ensure that the server can still operate stably under high load.

[0084] The process of constructing a monitoring indicator system includes the following steps:

[0085] Metrics Categorization and Organization: All extracted metrics are categorized according to hardware components, such as CPU metrics, memory metrics, disk metrics, network metrics, etc. Each category contains multiple specific metrics, forming a hierarchical metric structure.

[0086] Establishing metric correlations: Analyzing the interrelationships between different metrics, such as how high CPU utilization can lead to increased memory access latency, and how busy disk I / O can affect network data transmission speed. By establishing metric correlation models, a more comprehensive understanding of the server hardware system's operating status can be achieved, allowing for the rapid identification of other potentially affected metrics and related components when a particular metric becomes abnormal.

[0087] Establish a complete system: Integrate the categorized and organized indicators and their threshold ranges to form a complete server hardware status monitoring indicator system. This system should be scalable so that new indicators can be easily added or the thresholds of existing indicators can be adjusted as server hardware is upgraded or business needs change.

[0088] S200: Deploys a data acquisition agent to collect server hardware status data in real time, compares real-time data with monitoring indicator systems, assesses the current operating status and failure risks of the server, and outputs diagnostic data.

[0089] In this implementation, a data acquisition agent is deployed to collect server hardware status data in real time. This data is then compared with the monitoring indicator system to assess the server's current operating status and failure risks, and diagnostic data is output. The specific process is as follows:

[0090] S201: Deploy the data acquisition agent and set its configuration; the configuration includes the data acquisition frequency, the list of acquisition indicators, and the data transmission target address;

[0091] S202: Continuously collect server hardware status data according to the data collection frequency;

[0092] S203: Establish a virtual buffer, receive server hardware status data, assign timestamps to the data, preprocess the received data, and store the data based on the timestamps;

[0093] S204: Extract real-time indicator data and compare the real-time indicator data with the corresponding threshold range in the monitoring indicator system;

[0094] S205: Based on the comparison results of various indicators, evaluate the overall operating status of the server hardware and output comprehensive status data;

[0095] S206: Obtain indicator change trends, predict server failure risks based on historical data and indicator change trends, and output failure risk data;

[0096] S207: Generate a current server diagnostic report based on comprehensive status data and fault risk data, and display it visually.

[0097] In the above process, selecting a suitable agent is crucial: Choose data acquisition agent software that is compatible with the server's operating system and hardware architecture. For example, for Linux servers, open-source options like Zabbix Agent or Prometheus Node Exporter can be chosen; for Windows servers, the built-in Windows performance monitor tool or third-party commercial software such as:

[0098] SolarWinds Server&Application Monitor.

[0099] Install and configure the agent: Install the selected data acquisition agent on each server that needs to be monitored and configure it accordingly. Configuration includes specifying the data acquisition frequency, the list of metrics to be collected, and the target address for data transmission (such as the IP address and port of the monitoring server). Ensure the agent can run stably according to the set parameters and accurately collect server hardware status data.

[0100] Testing the agent's functionality: After the agent is installed and configured, perform a functional test to verify whether it can correctly collect data and send it to the designated monitoring server. The agent's working status can be confirmed by checking whether the data received on the monitoring server matches the server's actual hardware status.

[0101] Real-time and continuous data collection of server hardware status data: The data collection agent continuously collects server hardware status data at a preset frequency (e.g., every minute, every 5 minutes, etc.). A virtual buffer is established to receive server hardware status data, assign timestamps to the data, and preprocess the received data, such as deleting blank data and classifying the data. The data is then stored based on the timestamps.

[0102] Data storage management involves storing the collected real-time data in a dedicated database for subsequent querying and analysis. Relational databases such as MySQL and Oracle can be chosen, as well as time-series databases such as InfluxDB and TimescaleDB. Time-series databases offer higher performance and efficiency in processing time-series data.

[0103] Data matching analysis compares each real-time collected indicator data with the corresponding threshold range in the monitoring indicator system. For example, it compares the real-time collected CPU temperature with a pre-set CPU temperature threshold range to determine whether the current CPU temperature is within the normal range.

[0104] The anomaly detection mechanism uses a threshold comparison method to compare the basic threshold, the early warning threshold, and the fault threshold. Based on the comparison, it determines the current status of the indicator data. If the current indicator data is within the basic threshold, it is considered normal; if it is within the early warning threshold, it is considered an early warning; and if it is within the fault threshold, it is considered a fault.

[0105] A comprehensive status assessment evaluates the overall operational status of the server hardware based on comparisons of various indicators. For example, if multiple key indicators (such as CPU utilization, memory usage, and disk I / O latency) simultaneously exceed threshold ranges, it indicates that the server may be under high load and faces a high risk of failure.

[0106] Failure risk prediction combines historical data and indicator trends to forecast server failure risks. For example, if the number of bad sectors on a disk shows a continuous upward trend over a period of time, even if the current number of bad sectors has not yet reached the threshold, it can be predicted that the disk may fail soon, allowing for proactive measures such as data backup and disk replacement.

[0107] Generate a diagnostic report. Based on the status assessment and fault risk analysis results, generate a detailed diagnostic report. The report includes basic server information, real-time data for various indicators, comparisons with thresholds, current operational status assessment conclusions, fault risk warnings, and recommended handling measures.

[0108] Data visualization displays diagnostic data through a graphical interface, enabling operations and maintenance personnel to understand the server's operational status more intuitively. For example, line charts can be used to show the trend of CPU utilization over time, and bar charts can be used to compare the remaining space of different disks.

[0109] S300: Set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

[0110] In this implementation, an operation and maintenance strategy library is set up to generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information. The specific process is as follows:

[0111] S301: Define strategy categories and configure strategy parameters, set up the operation and maintenance strategy library, and optimize the operation and maintenance strategy library;

[0112] S302: Establish a decision generation engine to generate operation and maintenance decisions based on diagnostic data and execute the decisions according to the process;

[0113] S303: Obtain execution progress information in real time and verify the execution results;

[0114] S304: Establish a feedback mechanism to collect information and feedback from operations and maintenance personnel during the implementation of operational and maintenance decisions.

[0115] In the above process, strategy classification and formulation are carried out: corresponding operation and maintenance strategies are formulated according to different fault types and server operating status. Operation and maintenance strategies can be divided into preventative strategies, corrective strategies, and emergency strategies. Preventative strategies aim to avoid faults through regular maintenance and optimization, such as regularly cleaning up disk space and updating server firmware; corrective strategies are used for repair and recovery after a fault occurs, such as replacing faulty hardware and fixing software errors; emergency strategies are temporary measures to ensure the basic functions of the server or data security in the event of serious faults or emergencies, such as activating a backup server and performing emergency data backups.

[0116] Policy parameter configuration: Set corresponding parameters and trigger conditions for each operation and maintenance policy. For example, for the disk space cleanup policy, set it to trigger the cleanup operation when the remaining disk space is less than 20%; for the CPU overheat protection policy, set it to automatically reduce the CPU frequency to reduce the temperature when the CPU temperature exceeds the threshold for 5 consecutive minutes.

[0117] Policy Review and Optimization: Regularly review and optimize the operation and maintenance policy library. Adjust policy parameters and triggering conditions based on actual operation and maintenance experience and new business needs to ensure the effectiveness and adaptability of the policies. For example, with the improvement of server hardware performance and changes in business load, it may be necessary to adjust the CPU utilization warning threshold and corresponding operation and maintenance policies.

[0118] Decision generation logic: A decision generation engine is established to automatically generate corresponding operation and maintenance decisions based on the fault type and severity in the diagnostic data and the rules in the operation and maintenance strategy library. For example, if the diagnostic data shows that a hard drive has failed and the data cannot be read, the decision engine will generate a decision to replace the hard drive and restore the data based on the hard drive failure handling strategy in the operation and maintenance strategy library.

[0119] Decision execution process: The generated operation and maintenance decisions are sent to the corresponding execution modules, which then execute the decisions according to a predetermined process. For example, for a decision to replace a hard drive, the execution module will first notify the operation and maintenance personnel to prepare a new hard drive, then guide them to perform the hard drive replacement operation, and finally perform data recovery and system verification after the replacement is completed.

[0120] Automated execution versus manual intervention: For simple and low-risk operational decisions, automated execution can be implemented to improve operational efficiency. Examples include automatically cleaning up temporary disk files and automatically restarting services experiencing brief failures. For complex or high-risk decisions, such as upgrading server operating systems or replacing core hardware, manual confirmation and execution by operations personnel are required after the decision is generated to ensure the security, continuity, and stability of the operation.

[0121] Execution progress tracking: During the execution of operational decisions, real-time progress information is obtained. For example, for data backup decisions, information such as the completion percentage of the backup task and the amount of data already backed up is obtained through the backup software's interface, and this information is displayed on the operational management interface, allowing operational personnel to understand the execution status at any time.

[0122] Execution Result Verification: After the decision is executed, the results are verified to ensure that the decision achieves the expected results. For example, after replacing the hard drive and restoring the data, data verification tools are used to verify the integrity, continuity, and stability of the data; after restarting the service, it is checked whether the service is running normally and can respond to business requests normally.

[0123] Feedback Collection: Establish a feedback mechanism to collect relevant information and feedback from operations and maintenance personnel during the implementation of operational and maintenance decisions. This includes, for example, problems encountered by operations and maintenance personnel during decision implementation and suggestions for improvement.

[0124] Corresponding to the aforementioned embodiments of the intelligent server operation and maintenance method, this application also provides embodiments of the intelligent server operation and maintenance system.

[0125] Figure 5 This is a block diagram of a server intelligent operation and maintenance system according to an exemplary embodiment. (Refer to...) Figure 5 The system may include: a monitoring indicator system construction module 401, a fault prediction and diagnosis module 402, and an operation and maintenance decision execution feedback module 403; wherein:

[0126] The monitoring indicator system construction module 401 is used to acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system.

[0127] The fault prediction and diagnosis module 402 is used to deploy a data acquisition agent, collect server hardware status data in real time, compare real-time data with the monitoring indicator system, assess the current operating status and fault risk of the server, and output diagnostic data.

[0128] The operation and maintenance decision execution feedback module 403 is used to set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

[0129] In this embodiment, the monitoring indicator system construction module 401 acquires server hardware status data, extracts indicator features, sets threshold ranges for each indicator, and constructs a monitoring indicator system. The fault prediction and diagnosis module 402 deploys a data acquisition agent to collect server hardware status data in real time, compares the real-time data with the monitoring indicator system, assesses the current operating status and fault risks of the server, and outputs diagnostic data. The operation and maintenance decision execution feedback module 403 sets up an operation and maintenance strategy library, generates and executes operation and maintenance decisions based on the diagnostic data, acquires execution progress and results in real time, and receives feedback information. By constructing a monitoring indicator system based on indicator features and evaluating it based on real-time server hardware status data to generate corresponding operation and maintenance decisions, the operating status and potential fault risks of the server can be identified more accurately, avoiding the subjectivity and error of manual judgment and improving the continuity and stability of operation and maintenance.

[0130] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0131] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0132] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the server intelligent operation and maintenance method described above. Figure 6 The diagram shown is a hardware structure diagram of any data processing-capable device where a server intelligent operation and maintenance system provided in an embodiment of the present invention is located, except... Figure 6 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0133] Accordingly, this application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the server intelligent operation and maintenance method described above. The computer-readable storage medium can be an internal storage unit of any data processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0134] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0135] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A server intelligent operation and maintenance method, characterized in that, Includes the following steps: Acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system; Deploy a data acquisition agent to collect server hardware status data in real time, compare real-time data with monitoring indicator system, assess the current operating status and failure risk of the server, and output diagnostic data; Set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

2. The intelligent server operation and maintenance method as described in claim 1, characterized in that, In the steps of acquiring server hardware status data, extracting indicator features, setting threshold ranges for each indicator, and constructing a monitoring indicator system: The server's management interface is used to collect multi-dimensional status data of the server hardware and preprocess the multi-dimensional status data. Extract static, dynamic, and correlation indicator features from multi-dimensional state data, and filter out redundant data in the indicator features; Each indicator characteristic has a set range of basic threshold, early warning threshold and fault threshold, and the thresholds are dynamically adjusted based on historical data; By integrating indicator data, threshold ranges, and data sources, a monitoring indicator system is constructed.

3. The intelligent server operation and maintenance method as described in claim 2, characterized in that, In the steps of collecting multi-dimensional status data of server hardware through the server's management interface and preprocessing the multi-dimensional status data: Clean the data, remove noise and outliers, identify missing data, and complete the data accordingly. Set a collection period. When there is a short-term data gap, use the average of data from adjacent periods to complete the data. When data is missing for a long period of time, cache the file and fill in the historical data.

4. The server intelligent operation and maintenance method as described in claim 1, characterized in that, In the steps of extracting static, dynamic, and correlated indicator features from multi-dimensional state data and filtering out redundant data in the indicator features, the process of extracting correlated indicator features is as follows: Establish an indicator correlation model to analyze the interrelationships between multiple static and dynamic indicators, and create an indicator combination, which serves as the feature of the correlation indicator.

5. The server intelligent operation and maintenance method as described in claim 1, characterized in that, In the steps of deploying a data acquisition agent to collect server hardware status data in real time, comparing real-time data with monitoring indicator systems, assessing the current operating status and failure risks of the server, and outputting diagnostic data: Deploy the data acquisition agent and configure its settings; The configuration includes data collection frequency, a list of collection indicators, and the target address for data transmission. Continuously collect server hardware status data based on the data collection frequency; Extract real-time indicator data and compare the real-time indicator data with the corresponding threshold range in the monitoring indicator system; Based on the comparison results of various indicators, the overall operating status of the server hardware is evaluated, and comprehensive status data is output. Obtain the trend of indicator changes, predict the server failure risk based on historical data and indicator change trends, and output failure risk data.

6. The server intelligent operation and maintenance method as described in claim 5, characterized in that, After the step of continuously collecting server hardware status data according to the data collection frequency: A virtual buffer is established to receive server hardware status data, assign timestamps to the data, preprocess the received data, and store the data based on the timestamps.

7. The intelligent server operation and maintenance method as described in claim 5, characterized in that, After obtaining the trend of indicator changes, predicting the server failure risk based on historical data and indicator change trends, and outputting the failure risk data: Based on comprehensive status data and fault risk data, a current server diagnostic report is generated and visualized.

8. The intelligent server operation and maintenance method as described in claim 1, characterized in that, In the steps of setting up an operation and maintenance strategy library, generating and executing operation and maintenance decisions based on diagnostic data, obtaining execution progress and results in real time, and receiving feedback information: Define strategy categories and configure strategy parameters, set up an operation and maintenance strategy library, and optimize the operation and maintenance strategy library; Establish a decision generation engine to generate operation and maintenance decisions based on diagnostic data, and execute the decisions according to the process; Establish a feedback mechanism to collect information and feedback from operations and maintenance personnel during the implementation of operations and maintenance decisions.

9. The intelligent server operation and maintenance method as described in claim 8, characterized in that, After establishing a decision generation engine, generating operational decisions based on diagnostic data, and executing those decisions according to the established process: Get execution progress information in real time and verify the execution results.

10. A server intelligent operation and maintenance system, applied to the server intelligent operation and maintenance method as described in claim 1, characterized in that, This includes a monitoring indicator system construction module, a fault prediction and diagnosis module, and an operation and maintenance decision execution feedback module; among which: The monitoring indicator system construction module is used to acquire server hardware status data, extract indicator features, set threshold ranges for each indicator, and construct a monitoring indicator system. The fault prediction and diagnosis module is used to deploy a data acquisition agent to collect server hardware status data in real time, compare real-time data with the monitoring indicator system, assess the current operating status and fault risk of the server, and output diagnostic data. The operation and maintenance decision execution feedback module is used to set up an operation and maintenance strategy library, generate and execute operation and maintenance decisions based on diagnostic data, obtain execution progress and results in real time, and receive feedback information.

Images