CN119883852B - Server performance intelligent monitoring method and system - Google Patents
Server performance intelligent monitoring method and system Download PDFInfo
- Publication number
- CN119883852B CN119883852B CN202510370279.0A CN202510370279A CN119883852B CN 119883852 B CN119883852 B CN 119883852B CN 202510370279 A CN202510370279 A CN 202510370279A CN 119883852 B CN119883852 B CN 119883852B
- Authority
- CN
- China
- Prior art keywords
- load
- performance
- data
- performance data
- server
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
- G06F11/3433—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment for load management
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3452—Performance evaluation by statistical analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Debugging And Monitoring (AREA)
Abstract
本发明公开了服务器性能智能监控方法及系统，涉及服务器性能技术领域，包括分析负载历史数据确定负载的区间范围和变化程度，为后续负载的区间范围的拆分提供可靠基础，兼顾了服务器负载的以往变化情况，从而对负载的详细区间范围与性能数据之间的关联关系进行准确描述。确认负载的不同部分区间下负载与性能历史数据的关联关系，得到负载‑性能的标准关联关系，在负载与性能历史数据的对应关系中，挑选出性能历史数据的正常值。将负载数据和性能数据输入到负载‑性能的标准关联关系中来确定偏差情况，分析偏差情况输出目标服务器的当前性能信息，提高服务器性能监控的准确性和适应性，有效的保证了服务器的高效安全运行。
The present invention discloses a method and system for intelligent monitoring of server performance, which relates to the technical field of server performance, including analyzing historical load data to determine the range and degree of change of the load, providing a reliable basis for the subsequent splitting of the range of the load, taking into account the previous changes in the server load, and accurately describing the relationship between the detailed range of the load and the performance data. Confirm the relationship between the load and the historical performance data in different partial intervals of the load, obtain the standard relationship between the load and the performance, and select the normal value of the historical performance data in the corresponding relationship between the load and the historical performance data. Input the load data and the performance data into the standard relationship between the load and the performance to determine the deviation, analyze the deviation and output the current performance information of the target server, improve the accuracy and adaptability of the server performance monitoring, and effectively ensure the efficient and safe operation of the server.
Description
Technical Field
The invention relates to the technical field of server performance, in particular to an intelligent monitoring method and system for server performance.
Background
The background technology of the intelligent monitoring scheme for the server performance mainly stems from the continuous improvement of the requirements of the information technology field on the stability and the reliability of the server. With the expansion of enterprise business and the acceleration of digital transformation, the server is used as a core infrastructure supporting various application systems, and the performance of the server directly influences the continuity of business and user experience. The traditional server management mode is often difficult to accurately find potential problems in real time, so that fault early warning and disposal are delayed.
Therefore, an intelligent monitoring scheme is developed, which combines data collection, data analysis and early warning mechanisms, and performs anomaly detection and performance trend prediction by using a machine learning algorithm through collecting key performance indexes such as load, response time, throughput and the like of a server in real time. The techniques not only improve the accuracy and efficiency of monitoring, but also realize comprehensive evaluation and optimization of the server performance. The intelligent monitoring scheme can timely find out performance bottlenecks and potential fault risks, provides powerful data support for operation and maintenance teams, and ensures stable operation of the server and continuity of service.
In the prior art, the performance of the server is monitored according to the parameter of the performance data, and the influence of different loads of the server on the performance is not considered, so that the accuracy and adaptability of the performance monitoring of the server are poor, and the safe operation of the server cannot be effectively ensured.
Therefore, how to improve the accuracy and adaptability of server performance monitoring is a technical problem to be solved at present.
Disclosure of Invention
The invention aims to solve the problem that the accuracy and the adaptability of server performance monitoring are poor because the influence of different load conditions is not considered in the prior art, and provides a server performance intelligent monitoring method which comprises the following steps of,
Collecting load historical data of a target server, analyzing the load historical data to determine a range and a change degree of a load, and splitting the range of the load according to the change degree to obtain a plurality of partial intervals of the load;
Collecting performance history data of a target server according to a plurality of partial intervals of a load, and confirming association relations between the load and the performance history data under different partial intervals of the load to obtain standard association relations of the load and the performance;
load data and performance data of a target server in a current period of time are obtained, and the load data and the performance data are input into a standard association relation of load-performance to determine deviation conditions;
and analyzing the deviation condition and outputting the current performance information of the target server so as to monitor the performance of the server.
In some embodiments of the application, the load history data is analyzed to determine the range of intervals and the degree of variation of the load, including,
Drawing a histogram of the load according to the load history data, marking a minimum value and a maximum value of the load on the histogram, and taking a range from the minimum value to the maximum value of the load as an interval range of the load;
Identifying a load gradual change section and a load jump section from load historical data, and respectively integrating the load gradual change section and the load jump section to obtain a load gradual change section record and a load jump section record;
Determining the occurrence frequencies of different ranges under the interval range of the load based on the load gradual change section record and the load jump section record, respectively determining the frequent weights of the load gradual change section record and the load jump section record under the different ranges respectively through the occurrence frequencies, differentiating the different ranges of the load gradual change section record and the load jump section record, and carrying out differential weighting by combining the frequent weights under the corresponding ranges to obtain the gradual change degree and the jump change degree;
the degree of change of the interval range of the load is determined by integrating the degree of change of the gradual change and the degree of change of the jump.
In some embodiments of the present application, the range of intervals of the load is split according to the degree of variation, resulting in a plurality of partial intervals of the load, including,
Calculating the discrete degree of the load historical data, determining a splitting quantity according to the discrete degree and the change degree of the interval range of the load, and correspondingly splitting the interval range of the load through the splitting quantity to obtain a plurality of partial intervals of the load.
In some embodiments of the present application, the association of the load with the performance history data in different partial intervals of the load is confirmed to obtain a standard association of the load-performance, including,
Classifying the performance data, determining the corresponding performance data types and the distribution range of each performance data under different part intervals of the load, calculating the sensitivity between the load and each performance data, and counting the distribution parameters of each performance data;
the distribution parameters of the performance data comprise mean, standard deviation, quartile, IQR, skewness and kurtosis;
setting a first Z score threshold according to the mean value and the standard deviation, setting a second Z score threshold according to the quartile and the IQR, and setting a third Z score threshold according to the skewness and the kurtosis;
Setting a fourth Z score threshold by combining the first Z score threshold, the second Z score threshold and the third Z score threshold;
and distinguishing normal values and abnormal values in the distribution range of each performance data through a fourth Z score threshold value, eliminating the abnormal values, and constructing a standard association relation of load-performance according to the normal values and the sensitivity between the load and each performance data in the distribution range of the performance data.
In some embodiments of the application, the standard association of load-performance is constructed from normal values in the distribution range of performance data and the sensitivity between load and each performance data, including,
Fitting the relation between the load and the distribution range of the performance data under different part intervals on the basis of the normal value in the distribution range of the performance data to obtain a regression model, and adjusting the regression coefficient in the regression model through the sensitivity between the load and each performance data to describe the standard association relation of the load and the performance.
In some embodiments of the application, the load data and the performance data are input into a standard association of load-performance to determine a bias condition, including,
Inputting load data and performance data into a corresponding regression model according to different partial intervals of the load data, counting deviation under each load partial interval, and drawing a deviation curve of each load partial interval;
And integrating the deviation curves of all the load part intervals, extracting the curve characteristics of all the deviation curves, and generating a deviation condition.
In some embodiments of the present application, analyzing the bias conditions to output current performance information of the target server includes,
Determining curve characteristics under each type of performance data, generating a matching coefficient under each type of performance data by integrating the curve characteristics, determining a basic performance index of each type of performance data according to parameters of each type of performance data, and determining a performance index by combining the basic performance index and the matching coefficient of each type of performance data;
and integrating the performance indexes under all the performance data categories to output the current performance information of the target server.
In some embodiments of the present application, the performance metrics under all performance data categories are integrated to output current performance information for the target server, including,
Integrating the performance indexes under all performance data categories to obtain the overall performance index of the target server;
and analyzing the performance indexes under all the performance data categories to obtain the performance status of each type of performance data, and taking the performance status, the performance indexes and the overall performance indexes of each type of performance data as the current performance information of the target server.
Correspondingly, the application also provides an intelligent monitoring system for server performance, which comprises,
The first module is used for collecting load historical data of the target server, analyzing the load historical data to determine a range and a change degree of a load, and splitting the range of the load according to the change degree to obtain a plurality of partial intervals of the load;
The second module is used for collecting performance history data of the target server according to a plurality of partial intervals of the load, confirming association relations between the load and the performance history data under different partial intervals of the load, and obtaining standard association relations of the load-performance;
The third module is used for acquiring load data and performance data of the target server in a current period of time, and inputting the load data and the performance data into a standard association relation of load-performance to determine a deviation condition;
and the fourth module is used for analyzing the deviation condition and outputting the current performance information of the target server so as to monitor the performance of the server.
Compared with the prior art, the invention has the beneficial effects that:
1. The load history data is analyzed to determine the interval range and the change degree of the load, a reliable basis is provided for splitting the interval range of the subsequent load, and the previous change condition of the load of the server is considered, so that the association relation between the detailed interval range of the load and the performance data is accurately described. And confirming the association relation between the load and the performance history data in different part intervals of the load to obtain a standard association relation between the load and the performance, selecting normal values of the performance history data in the corresponding relation between the load and the performance history data, constructing the standard association relation between the load and the performance, and ensuring the accuracy of the description of the standard association relation between the load and each performance data in different load intervals.
2. The load data and the performance data are input into the standard association relation of load-performance to determine the deviation condition, the current performance information of the target server is output by analyzing the deviation condition, the accuracy and the adaptability of the performance monitoring of the server are improved, and the efficient and safe operation of the server is effectively ensured.
Drawings
FIG. 1 is a schematic flow chart of a method for intelligently monitoring server performance;
fig. 2 is a schematic structural diagram of the intelligent monitoring system for server performance according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Referring to fig. 1, the server performance intelligent monitoring method includes the steps of,
Step S101, collecting load historical data of a target server, analyzing the load historical data to determine a range and a change degree of a load, and splitting the range of the load according to the change degree to obtain a plurality of partial intervals of the load.
In this embodiment, system commands (e.g., uptime, top, sar, etc.) or specialized monitoring tools (e.g., zabbix, nagios, prometheus, etc.) are used to collect load history data for the target server. Ensuring that the selected tool is able to provide accurate, comprehensive load data including, but not limited to, key metrics such as CPU utilization, memory utilization, network bandwidth, etc. The collected load history data is stored in a secure, reliable database or file system. The data is subjected to necessary sort work, such as deduplication, formatting, etc., to ensure the integrity and consistency of the data. Preprocessing the collected load history data, including data cleaning (such as abnormal value removal, missing value removal and the like) and data normalization. The preprocessed data can be ensured to accurately reflect the load condition of the server. And analyzing the preprocessed load data by using a statistical method (such as a histogram, a box diagram and the like) to determine the interval range of the load. In the prior art, the load data is divided into three sections of low load, medium load and high load according to the distribution condition of the load data, but the load section of the fuzzy general system is not suitable for analyzing the specific relation between the load and the performance, so the load section is required to be split according to the conventional change condition of the load of a server, and the relation between the load and the performance under the specific section of the load is analyzed.
In some embodiments of the application, the load history data is analyzed to determine the range of intervals and the degree of variation of the load, including,
Drawing a histogram of the load according to the load history data, marking a minimum value and a maximum value of the load on the histogram, and taking a range from the minimum value to the maximum value of the load as an interval range of the load;
Identifying a load gradual change section and a load jump section from load historical data, and respectively integrating the load gradual change section and the load jump section to obtain a load gradual change section record and a load jump section record;
Determining the occurrence frequencies of different ranges under the interval range of the load based on the load gradual change section record and the load jump section record, respectively determining the frequent weights of the load gradual change section record and the load jump section record under the different ranges respectively through the occurrence frequencies, differentiating the different ranges of the load gradual change section record and the load jump section record, and carrying out differential weighting by combining the frequent weights under the corresponding ranges to obtain the gradual change degree and the jump change degree;
the degree of change of the interval range of the load is determined by integrating the degree of change of the gradual change and the degree of change of the jump.
In this embodiment, the interval range of the load is the range between the minimum value and the maximum value of the load on the histogram, and is all possible ranges of the load. There are two modes of load variation of the server, one is gradual and the other is direct jump. Gradual changes, where the load continues to increase or decrease over a period of time, are generally more smooth and are suitable for counting frequency by interval division to reflect the overall trend of the load. Direct jump-the load changes significantly in a short time, which is usually severe, and is suitable for capturing the fluctuation of the load by statistics of specific values. And integrating all the load gradual change sections and all the load jump sections respectively to obtain a load gradual change section record and a load jump section record. On the premise of two modes, the occurrence frequencies of different ranges under the interval range of the load are calculated, frequent weights are given, and when the difference is calculated, the difference can be weighted according to the occurrence frequencies of the load interval. The frequently occurring load interval should be given higher weight in the differential calculation so as to reflect the influence of the load interval on the whole load fluctuation more accurately, and the change degree of the load is reflected by combining the two modes.
In some embodiments of the present application, the range of intervals of the load is split according to the degree of variation, resulting in a plurality of partial intervals of the load, including,
Calculating the discrete degree of the load historical data, determining a splitting quantity according to the discrete degree and the change degree of the interval range of the load, and correspondingly splitting the interval range of the load through the splitting quantity to obtain a plurality of partial intervals of the load.
In this embodiment, the discrete degree of the load history data also illustrates the change condition of the load, and a splitting number is determined by combining the discrete degree and the change degree (such as weighted summation) of the interval range of the load, and the higher the discrete degree and the change degree, the larger the splitting number, so that the relationship between the load and the performance data can be accurately analyzed. Too many intervals may increase the complexity of the analysis, while too few intervals may not accurately reflect the relationship between load and performance, so a reasonable number of splits is determined.
Step S102, collecting performance history data of the target server according to a plurality of partial intervals of the load, and confirming association relations between the load and the performance history data under different partial intervals of the load to obtain standard association relations of the load and the performance.
In this embodiment, an appropriate performance index is selected according to an application scenario and a service requirement of the server. The performance index should be able to comprehensively reflect the performance status of the server, such as response time, throughput, number of concurrent users, etc. The usage monitoring tool or system commands collection of performance history data in terms of multiple fractional spans of the load. Ensuring that the collected performance data is consistent with the load data over time for subsequent correlation analysis.
Server performance is embodied in multiple aspects or dimensions, each of which directly reflects a different performance condition of the server. The following is a detailed summary and description of these aspects or dimensions:
1. CPU performance
The method has the advantages of high processing speed and concurrent processing capacity of the server.
Performance conditions:
Processing speed-CPU is the core component of the server, responsible for executing various computing tasks. The high-performance CPU can execute the calculation task faster, and the response speed and performance of the server are improved. For example, in application scenarios where data analysis, scientific computation, etc. require a large amount of computation, the performance of the CPU is critical.
Concurrent processing capability modern CPUs typically have multi-core designs capable of processing multiple tasks simultaneously. The multi-core CPU can enhance the concurrent processing capacity of the server, so that the server can process more user requests or tasks at the same time. This is important for running large databases, virtualized environments, or highly loaded applications.
2. Memory performance
The method has the advantages of high task processing capacity, high system stability and high database performance.
Performance conditions:
The task processing capability is that the memory is used for temporarily storing data generated in the CPU, and enough memory can enable the server to process more data and requests, so that the response speed is improved. Meanwhile, the memory also supports more programs to run simultaneously, so that the switching time between the programs is reduced, and the response speed of the system is improved.
The system stability, namely the larger memory capacity can relieve the problem of shortage of system resources and provide more sufficient running space, thereby improving the stability and reliability of the system.
Database performance, namely, the memory is used as a cache area of the database, and has important influence on the performance of the database. The larger memory capacity can provide more cache space, reduce the access times of the disk and greatly improve the query and writing speeds of the database.
3. Disk performance
The method has the advantages of high data storage and reading speed of the server and high overall system efficiency.
Performance conditions:
Data storage and read speeds high performance hard disks (e.g., SSDs or NVMe SSDs) provide data read and write speeds much higher than traditional mechanical hard disks (HDDs). This directly relates to the response time of the application and the speed of the data processing. For example, database queries, log records, file transfers, etc. may be significantly accelerated.
The faster disk read-write speed means shorter starting time of the operating system and the application program, and the overall efficiency of the system is higher. Meanwhile, the high-performance hard disk can also support higher IOPS (input/output operation times per second), and stable operation can be ensured under high load.
4. Network performance
The method and the device have the advantages of data transmission speed and concurrent request processing capability of the server.
Performance conditions:
Data transmission speed-the network bandwidth determines the speed of data transmission in the network. The high network bandwidth can reduce data transmission delay and improve the efficiency of distributed computation and remote access.
Concurrent request processing capability-in a high concurrency scenario, a server needs to process a large number of requests at the same time. Sufficient network bandwidth can support more concurrent connections, ensuring that the server can handle these requests in a timely manner.
5. Reliability and stability
The fault tolerance capability and the long-time stable operation capability of the server are embodied.
Performance conditions:
fault tolerance-high quality server hardware and software design can improve the fault tolerance of the server. For example, redundant power supplies and fans, predictable hard disk and fan failures, RAID systems, and the like are common techniques for improving server reliability.
The long-time stable operation capability is that the stability and the reliability of the server are also reflected in that the server can stably operate for a long time without faults or performance degradation. This is critical for application scenarios (e.g., websites, databases, etc.) where continuous service is required.
In this embodiment, performance history data of the target server is collected according to a plurality of partial intervals of the load, and performance history data of the target server corresponding to each interval is collected based on the partial intervals of the load.
In some embodiments of the present application, the association of the load with the performance history data in different partial intervals of the load is confirmed to obtain a standard association of the load-performance, including,
Classifying the performance data, determining the corresponding performance data types and the distribution range of each performance data under different part intervals of the load, calculating the sensitivity between the load and each performance data, and counting the distribution parameters of each performance data;
the distribution parameters of the performance data comprise mean, standard deviation, quartile, IQR, skewness and kurtosis;
setting a first Z score threshold according to the mean value and the standard deviation, setting a second Z score threshold according to the quartile and the IQR, and setting a third Z score threshold according to the skewness and the kurtosis;
Setting a fourth Z score threshold by combining the first Z score threshold, the second Z score threshold and the third Z score threshold;
and distinguishing normal values and abnormal values in the distribution range of each performance data through a fourth Z score threshold value, eliminating the abnormal values, and constructing a standard association relation of load-performance according to the normal values and the sensitivity between the load and each performance data in the distribution range of the performance data.
In the present embodiment, the sensitivity between the load and each performance data is calculated for the distribution range of each performance data (the distribution range of the parameter value size), where the sensitivity can be calculated by the pearson correlation coefficient or the like. To accurately describe the relationship between load and performance, outlier data (errors, outlier performance, etc.) in the performance data is removed. The identification of normal data and abnormal data is carried out by adopting a Z-score method, the Z-score of each data is calculated, and a Z-score threshold value is compared to judge whether the data is abnormal data.
In this embodiment, the distribution parameters of the performance data include mean, standard deviation, quartile, IQR, skewness, and kurtosis, and specifically include the following:
Mean value is the most common method, which can help us to understand the central trend of data. In the Z-score method, the mean is used to calculate the difference between each data point and the mean.
The median is another central trend indicator of the data set, which is the value of the data in the middle after the data are arranged according to the order of the sizes. The median is not affected by the extreme value, and when the extreme value exists in the data set, the median can more accurately reflect the central position of the data than the mean value.
Mode is the value in the dataset that occurs most frequently. Mode can help us understand the most common case when analyzing data distribution.
Quartile the quartile can divide the data into four parts, each part containing 25% of the data points. The distance between the first quartile (Q1) and the third quartile (Q3) is referred to as the quartile range (IQR), which is an indicator of how discrete the data is. The quartile can help us quickly identify outliers, which are generally considered to be data points that fall outside of Q1-1.5IQR and Q3+1.5 IQR.
Skewness-skewness describes the symmetry of the data distribution. Positive bias means that the data distribution is right biased, i.e. larger value, and negative bias means that the data distribution is left biased, i.e. smaller value. Skewness can affect the setting of the Z-score threshold because data distributions with greater skewness may be more prone to extreme values.
Kurtosis-kurtosis describes the degree of spike in the data distribution. Kurtosis high indicates that the data is concentrated near the mean and kurtosis low indicates that the data is more diffuse. Kurtosis may also affect the setting of the Z-score threshold, as a data distribution with higher kurtosis may more easily produce data points that are close to the mean.
Based on the mean and standard deviation, in the case of data approaching normal distribution, the mean and standard deviation can be used to calculate the Z-score and a fixed threshold (e.g., 3) can be set to identify outliers. However, if the data distribution deviates from the normal distribution or the volatility is large, the threshold value may need to be adjusted.
Regarding quartiles and IQR, for data that is more fluctuating or that deviates from normal distribution, quartiles and IQR may be used to assist in setting the Z-score threshold. For example, the IQR of the data may be calculated and Q1-1.5IQR and Q3+1.5IQR may be used as boundaries for outliers. The threshold value of the Z-score is then adjusted based on these boundaries.
Combining skewness and kurtosis the skewness and kurtosis of the data may also be combined when considering setting the Z-score threshold. If the data skewness is greater or kurtosis is higher, it may be desirable to set a looser threshold to avoid misjudging normal fluctuations as anomalies.
In this embodiment, a fourth Z-score threshold is set in combination with the first Z-score threshold, the second Z-score threshold, and the third Z-score threshold, and the calculation formula is as follows:
;
Wherein, the Is the first to loadCorresponding under each partial intervalA fourth Z-score threshold for the performance data, 、 、 The contribution weights of the first Z-score threshold, the second Z-score threshold and the third Z-score threshold respectively, 、 、 Respectively the first of the loadsCorresponding under each partial intervalA first Z-score threshold, a second Z-score threshold, a third Z-score threshold of performance data,AndRespectively is 、 、 The maximum value and the minimum value of the three,Is the first to loadCorresponding under each partial intervalA first constant of performance data;
In the present embodiment of the present invention, The average of the maximum value and the minimum value of the influence quantity of the first Z fraction threshold value, the second Z fraction threshold value and the third Z fraction threshold value is corrected to the average value of the three values,To balance the size of the correction function.
In some embodiments of the application, the standard association of load-performance is constructed from normal values in the distribution range of performance data and the sensitivity between load and each performance data, including,
Fitting the relation between the load and the distribution range of the performance data under different part intervals on the basis of the normal value in the distribution range of the performance data to obtain a regression model, and adjusting the regression coefficient in the regression model through the sensitivity between the load and each performance data to describe the standard association relation of the load and the performance.
In this embodiment, an appropriate regression model (such as linear regression, polynomial regression, etc.) is selected to fit the relationship between the load and the distribution range of the performance data in different partial intervals on the basis of the normal values in the distribution range of the performance data. And training the model by using the collected data to obtain a preliminary regression coefficient. The sensitivity coefficient reflects the extent to which load changes affect performance data changes. If the sensitivity coefficient is larger, the influence of the load on the performance data is obvious, otherwise, the influence is smaller. And adjusting the regression coefficient in the regression model according to the result of the sensitivity analysis. If the sensitivity of a certain performance data to the load is higher, the regression coefficient corresponding to the performance data can be appropriately increased to enhance the influence of the load on the performance data. Conversely, if the sensitivity of a certain performance data to the load is low, the regression coefficient corresponding to the performance data may be appropriately reduced to reduce the influence of the load on the performance data. And verifying the adjusted regression model by using the verification data set, and evaluating the prediction accuracy and stability of the model. If the model is not well behaved, repeating the steps to perform model optimization until a satisfactory model is obtained, and describing the standard association relationship of load-performance under different load intervals through different regression models.
Step S103, load data and performance data of the target server in a current period of time are obtained, and the load data and the performance data are input into a standard association relation of load-performance to determine deviation conditions.
In the embodiment, the real-time data collection is that a monitoring tool is used for acquiring load data and performance data of the target server in real time in a current period of time. And recording the acquired data and storing the data in a safe and reliable database or file system. The data record should include both load data and performance data for subsequent correlation analysis and monitoring. And inputting the load data and the performance data into the corresponding regression model, and counting the deviation condition of the performance data.
In some embodiments of the application, the load data and the performance data are input into a standard association of load-performance to determine a bias condition, including,
Inputting load data and performance data into a corresponding regression model according to different partial intervals of the load data, counting deviation under each load partial interval, and drawing a deviation curve of each load partial interval;
And integrating the deviation curves of all the load part intervals, extracting the curve characteristics of all the deviation curves, and generating a deviation condition.
In this embodiment, the performance deviation is the difference between the predicted value and the actual value corresponding to different loads in the regression model, and the smaller the deviation, the better the actual load is matched with the performance, and the matching degree between the relationship between the actual load and the performance and the standard association relationship between the load and the performance is described. The curve characteristics of the deviation curve include slope, extremum, smoothness, etc.
And step S104, analyzing the deviation condition and outputting the current performance information of the target server, so as to realize the monitoring of the server performance.
In this embodiment, the current performance of the target server is analyzed by combining the parameter size and the deviation condition of the performance data, and the current performance is comprehensively analyzed from the parameter size and the load-performance relationship matching condition of the performance data.
In some embodiments of the present application, analyzing the bias conditions to output current performance information of the target server includes,
Determining curve characteristics under each type of performance data, generating a matching coefficient under each type of performance data by integrating the curve characteristics, determining a basic performance index of each type of performance data according to parameters of each type of performance data, and determining a performance index by combining the basic performance index and the matching coefficient of each type of performance data;
and integrating the performance indexes under all the performance data categories to output the current performance information of the target server.
In some embodiments of the present application, the performance metrics under all performance data categories are integrated to output current performance information for the target server, including,
Integrating the performance indexes under all performance data categories to obtain the overall performance index of the target server;
and analyzing the performance indexes under all the performance data categories to obtain the performance status of each type of performance data, and taking the performance status, the performance indexes and the overall performance indexes of each type of performance data as the current performance information of the target server.
In this embodiment, all the curve features are integrated (weighted sum after normalization, mapping to obtain a matching coefficient), so as to obtain a matching coefficient between the performance data and the load, and the parameter size of the index data of each type of performance is evaluated to obtain a basic performance index (the quality of performance parameter representation), and the performance indexes under all the performance data types are integrated to output the current performance information of the target server.
;
Wherein, the For the current performance information of the target server (total server performance),For the number of categories of performance data,Is the firstThe matching coefficients of the performance data are such that,Is the firstBasic performance indicators of the performance data,Is thatA number of performance data categories less than the corresponding threshold,For the conversion coefficient of the performance difference,Is thatA first of less than a corresponding thresholdThe matching coefficients of the performance data are such that,Is thatA first of less than a corresponding thresholdBasic performance indicators of the performance data,In order to set the constant value of the preset value,Representation ofA correction of the sum of the performances of all the performance data smaller than the sum of the corresponding thresholds,Is to balance the correction function.
Correspondingly, the application also provides an intelligent monitoring system for server performance, as shown in figure 2, comprising,
The first module is used for collecting load historical data of the target server, analyzing the load historical data to determine a range and a change degree of a load, and splitting the range of the load according to the change degree to obtain a plurality of partial intervals of the load;
The second module is used for collecting performance history data of the target server according to a plurality of partial intervals of the load, confirming association relations between the load and the performance history data under different partial intervals of the load, and obtaining standard association relations of the load-performance;
The third module is used for acquiring load data and performance data of the target server in a current period of time, and inputting the load data and the performance data into a standard association relation of load-performance to determine a deviation condition;
and the fourth module is used for analyzing the deviation condition and outputting the current performance information of the target server so as to monitor the performance of the server.
Compared with the prior art, the invention has the beneficial effects that:
1. The load history data is analyzed to determine the interval range and the change degree of the load, a reliable basis is provided for splitting the interval range of the subsequent load, and the previous change condition of the load of the server is considered, so that the association relation between the detailed interval range of the load and the performance data is accurately described. And confirming the association relation between the load and the performance history data in different part intervals of the load to obtain a standard association relation between the load and the performance, selecting normal values of the performance history data in the corresponding relation between the load and the performance history data, constructing the standard association relation between the load and the performance, and ensuring the accuracy of the description of the standard association relation between the load and each performance data in different load intervals.
2. The load data and the performance data are input into the standard association relation of load-performance to determine the deviation condition, the current performance information of the target server is output by analyzing the deviation condition, the accuracy and the adaptability of the performance monitoring of the server are improved, and the efficient and safe operation of the server is effectively ensured.
From the above description of the embodiments, it will be clear to those skilled in the art that the present invention may be implemented in hardware, or may be implemented by means of software plus necessary general hardware platforms. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.), and includes several instructions for causing a computer device (may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective implementation scenario of the present invention.
Those skilled in the art will appreciate that the drawing is merely a schematic illustration of a preferred implementation scenario and that the modules or flows in the drawing are not necessarily required to practice the invention.
Those skilled in the art will appreciate that modules in an apparatus in an implementation scenario may be distributed in an apparatus in an implementation scenario according to an implementation scenario description, or that corresponding changes may be located in one or more apparatuses different from the implementation scenario. The modules of the implementation scenario may be combined into one module, or may be further split into a plurality of sub-modules.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (8)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510370279.0A CN119883852B (en) | 2025-03-27 | 2025-03-27 | Server performance intelligent monitoring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510370279.0A CN119883852B (en) | 2025-03-27 | 2025-03-27 | Server performance intelligent monitoring method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119883852A CN119883852A (en) | 2025-04-25 |
| CN119883852B true CN119883852B (en) | 2025-07-08 |
Family
ID=95424707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510370279.0A Active CN119883852B (en) | 2025-03-27 | 2025-03-27 | Server performance intelligent monitoring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119883852B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354092A (en) * | 2015-11-19 | 2016-02-24 | 东软集团股份有限公司 | Method, device and system for predicting application performance risk |
| CN117311984A (en) * | 2023-11-03 | 2023-12-29 | 北京创璞科技有限公司 | Method and system for balancing server load based on comparison service |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11244410B2 (en) * | 2020-02-24 | 2022-02-08 | Leading Edge Power Solutions, Llc | Technologies for dynamically dispatching generator power |
| CN118585397A (en) * | 2024-05-29 | 2024-09-03 | 深圳市灏鑫达电子科技有限公司 | Solid state drive performance monitoring method and system |
| CN119421175A (en) * | 2024-10-15 | 2025-02-11 | 上海创蓝云智信息科技股份有限公司 | A real-time SMS communication link verification and performance monitoring method |
-
2025
- 2025-03-27 CN CN202510370279.0A patent/CN119883852B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354092A (en) * | 2015-11-19 | 2016-02-24 | 东软集团股份有限公司 | Method, device and system for predicting application performance risk |
| CN117311984A (en) * | 2023-11-03 | 2023-12-29 | 北京创璞科技有限公司 | Method and system for balancing server load based on comparison service |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119883852A (en) | 2025-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110413227B (en) | Method and system for predicting remaining service life of hard disk device on line | |
| JP5936509B2 (en) | Program, analysis method, and information processing apparatus | |
| US7502971B2 (en) | Determining a recurrent problem of a computer resource using signatures | |
| US8078913B2 (en) | Automated identification of performance crisis | |
| US9921936B2 (en) | Method and system for IT resources performance analysis | |
| CN111984499A (en) | Fault detection method and device for big data cluster | |
| CN119179598B (en) | Solid state disk fault prediction method and system based on artificial intelligence | |
| US6681309B2 (en) | Method and apparatus for measuring and optimizing spatial segmentation of electronic storage workloads | |
| CN118761745B (en) | OA collaborative workflow optimization method applied to enterprise | |
| CN104516808B (en) | Data prediction device and method | |
| WO2012142144A2 (en) | Assessing application performance with an operational index | |
| Liu et al. | Multi-task hierarchical classification for disk failure prediction in online service systems | |
| CN118673396A (en) | Big data platform operation and maintenance management system based on artificial intelligence | |
| CN118673087A (en) | Bank data warehouse construction method, system, equipment and storage medium | |
| Wildani et al. | Efficiently identifying working sets in block i/o streams | |
| CN118331822A (en) | Abnormal information detection method and device, storage medium and electronic device | |
| CN114860540B (en) | Cloud data center server health evaluation method | |
| CN118733367B (en) | Disk performance analysis method based on high-density storage | |
| CN118917390B (en) | Service knowledge base management system and method based on knowledge big model | |
| CN119883852B (en) | Server performance intelligent monitoring method and system | |
| CN118708593A (en) | A data processing method and system in a high-concurrency scenario of a data center | |
| CN118689887A (en) | Data consistency detection and repair method, device and electronic equipment | |
| WO2016013099A1 (en) | Feature data management system and feature data management method | |
| CN112749035A (en) | Anomaly detection method, device and computer readable medium | |
| CN111158974B (en) | A hardware-aware CPU energy consumption calculation method for cloud servers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |