Performance monitoring method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of computers, and discloses a performance monitoring method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining historical performance data of a storage system and configuration change information of the storage system; for any performance index in the historical performance data, determining a performance identification result corresponding to the performance index according to the performance index and the configuration change information of the storage system; determining a dynamic threshold value corresponding to each performance index according to the performance identification result corresponding to each performance index of the storage system in the preset period; and determining a performance monitoring result of the storage system according to the current performance index of the storage system and the corresponding dynamic threshold value. According to the method provided by the scheme, dynamic setting of the performance index threshold is carried out through the configuration change information adapted to the storage system, the situation that normal performance fluctuation is mistakenly considered to be abnormal due to the configuration change of the storage system is avoided, then false alarm is avoided, and the accuracy of a performance monitoring result is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a performance monitoring method, device, electronic device and storage medium. Background Art

[0002] At present, with the development of technologies such as big data, storage systems have been widely used, and the performance requirements for storage systems are gradually increasing. Therefore, how to monitor the performance of storage systems has become a key research topic.

[0003] In the related art, fixed thresholds of several performance indicators of the storage system are usually preset. When any performance indicator exceeds the corresponding fixed threshold, performance alarm information of the storage system is generated to remind relevant technical personnel to maintain the performance of the storage system.

[0004] However, as storage systems grow in size and complexity, performance monitoring based on the above approach can easily misinterpret normal performance fluctuations as abnormalities, causing false alarms and reducing the accuracy of performance monitoring results. Summary of the invention

[0005] The present application provides a performance monitoring method, device, electronic device and storage medium to address the defects of related technologies in improving the accuracy of storage system performance monitoring results.

[0006] The first aspect of the present application provides a performance monitoring method, comprising:

[0007] Acquire historical performance data of a storage system and configuration change information of the storage system;

[0008] For any performance indicator in the historical performance data, determine a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system;

[0009] Determine a dynamic threshold value corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period;

[0010] A performance monitoring result of the storage system is determined according to a current performance indicator of the storage system and a corresponding dynamic threshold.

[0011] In an optional implementation, for any performance indicator in the historical performance data, determining a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system includes:

[0012] Determining several change factors of the storage system according to the configuration change information of the storage system;

[0013] For any performance indicator in the historical performance data, selecting a key change factor that affects the performance indicator from among the several change factors;

[0014] Acquire a plurality of performance identification models, wherein the performance identification models are constructed based on a machine learning algorithm and historical operation data of the storage system;

[0015] For any of the performance identification models, the degree of change of the key change factor and the performance index are input into the performance identification model, so as to identify the normal fluctuation of the performance index based on the performance identification model and the degree of change of the key change factor and the performance index, and obtain a model output result;

[0016] Allocating a weight coefficient to each of the performance identification models according to the confidence of the model output result of each of the performance identification models;

[0017] Determining a performance identification result corresponding to the performance indicator according to the model output results of each of the performance identification models and the weight coefficients of each of the performance identification models;

[0018] The change factors of the storage system include at least hardware change, service change and load pressure adjustment.

[0019] In an optional implementation, determining the performance identification result corresponding to the performance indicator according to the model output results of each of the performance identification models and the weight coefficient of each of the performance identification models includes:

[0020] The performance identification result corresponding to the performance indicator is determined based on the following formula:

[0021]

[0022] Among them, f(x) represents the performance identification result corresponding to the performance indicator x, n represents the number of performance identification models, and w i represents the weight coefficient of performance identification model i, h i (x) represents the model output result of performance identification model i for performance indicator x, and b represents the variation factor bias.

[0023] In an optional implementation, determining the dynamic threshold corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period includes:

[0024] For any of the performance indicators, according to the performance identification results corresponding to the performance indicator within a preset period, determine the average performance identification result and the standard deviation of the performance identification result of the performance indicator;

[0025] A dynamic threshold corresponding to the performance indicator is determined according to an average performance identification result and a standard deviation of the performance identification result of the performance indicator.

[0026] In an optional implementation, determining the dynamic threshold corresponding to the performance indicator according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator includes:

[0027] Based on the following formula, the dynamic threshold corresponding to the performance indicator is determined according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator:

[0028] T=u+k·σ

[0029] Wherein, T represents the dynamic threshold corresponding to the performance indicator, u represents the average performance identification result of the performance indicator, k represents the adjustment coefficient, and σ represents the standard deviation of the performance identification result of the performance indicator.

[0030] In an optional implementation, determining the performance monitoring result of the storage system according to the current performance indicator of the storage system and the corresponding dynamic threshold includes:

[0031] Determine the current abnormal performance indicator according to the magnitude relationship between the current performance indicator of the storage system and the corresponding dynamic threshold;

[0032] According to the current abnormal performance indicator, performance alarm information is generated to obtain the performance monitoring result of the storage system.

[0033] In an optional embodiment, the method further includes:

[0034] Obtain performance false alarm information from storage system users;

[0035] According to the performance false alarm information, the dynamic threshold corresponding to each performance indicator is optimized.

[0036] A second aspect of the present application provides a performance monitoring device, comprising:

[0037] An acquisition module, used to acquire historical performance data of a storage system and configuration change information of the storage system;

[0038] A first determination module is used to determine, for any performance indicator in the historical performance data, a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system;

[0039] A second determination module, configured to determine a dynamic threshold corresponding to each performance indicator according to a performance identification result corresponding to each performance indicator of the storage system within a preset period;

[0040] The monitoring module is used to determine the performance monitoring result of the storage system according to the current performance index of the storage system and the corresponding dynamic threshold.

[0041] In an optional implementation manner, the first determining module is specifically configured to:

[0042] Determining several change factors of the storage system according to the configuration change information of the storage system;

[0043] For any performance indicator in the historical performance data, selecting a key change factor that affects the performance indicator from among the several change factors;

[0044] Acquire a plurality of performance identification models, wherein the performance identification models are constructed based on a machine learning algorithm and historical operation data of the storage system;

[0045] For any of the performance identification models, the degree of change of the key change factor and the performance index are input into the performance identification model, so as to identify the normal fluctuation of the performance index based on the performance identification model and the degree of change of the key change factor and the performance index, and obtain a model output result;

[0046] Allocating a weight coefficient to each of the performance identification models according to the confidence of the model output result of each of the performance identification models;

[0047] Determining a performance identification result corresponding to the performance indicator according to the model output results of each of the performance identification models and the weight coefficients of each of the performance identification models;

[0048] The change factors of the storage system include at least hardware change, service change and load pressure adjustment.

[0049] In an optional implementation manner, the first determining module is specifically configured to:

[0050] The performance identification result corresponding to the performance indicator is determined based on the following formula:

[0051]

[0052] Among them, f(x) represents the performance identification result corresponding to the performance indicator x, n represents the number of performance identification models, and w i represents the weight coefficient of performance identification model i, h i (x) represents the model output result of performance identification model i for performance indicator x, and b represents the variation factor bias.

[0053] In an optional implementation manner, the second determining module is specifically configured to:

[0054] For any of the performance indicators, according to the performance identification results corresponding to the performance indicator within a preset period, determine the average performance identification result and the standard deviation of the performance identification result of the performance indicator;

[0055] A dynamic threshold corresponding to the performance indicator is determined according to an average performance identification result and a standard deviation of the performance identification result of the performance indicator.

[0056] In an optional implementation manner, the second determining module is specifically configured to:

[0057] Based on the following formula, the dynamic threshold corresponding to the performance indicator is determined according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator:

[0058] T=u+k·σ

[0059] Wherein, T represents the dynamic threshold corresponding to the performance indicator, u represents the average performance identification result of the performance indicator, k represents the adjustment coefficient, and σ represents the standard deviation of the performance identification result of the performance indicator.

[0060] In an optional implementation, the monitoring module is specifically used to:

[0061] Determine the current abnormal performance indicator according to the magnitude relationship between the current performance indicator of the storage system and the corresponding dynamic threshold;

[0062] According to the current abnormal performance indicator, performance alarm information is generated to obtain the performance monitoring result of the storage system.

[0063] In an optional embodiment, the device further comprises:

[0064] The optimization module is used to obtain performance false alarm information fed back by users of the storage system; and optimize the dynamic threshold corresponding to each of the performance indicators according to the performance false alarm information.

[0065] A third aspect of the present application provides an electronic device, comprising: at least one processor and a memory;

[0066] The memory stores computer-executable instructions;

[0067] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method described in the first aspect and various possible designs of the first aspect.

[0068] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the method described in the first aspect and various possible designs of the first aspect are implemented.

[0069] The fifth aspect of the present application provides a computer program product, including computer instructions, which are used to enable a computer to execute the method described in the first aspect and various possible designs of the first aspect.

[0070] The technical solution of this application has the following advantages:

[0071] The present application provides a performance monitoring method, device, electronic device and storage medium, the method comprising: obtaining historical performance data of a storage system and configuration change information of the storage system; for any performance indicator in the historical performance data, determining the performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system; determining the dynamic threshold corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period; determining the performance monitoring result of the storage system according to the current performance indicator of the storage system and the corresponding dynamic threshold. The method provided by the above scheme dynamically sets the performance indicator threshold by adapting to the configuration change information of the storage system, thereby avoiding mistaking normal performance fluctuations for abnormalities due to changes in the storage system configuration, thereby avoiding false alarms and improving the accuracy of the performance monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the following is a brief introduction to the drawings required for use in the embodiments or the related technical descriptions. Obviously, the drawings described below are some embodiments of the present application, and a person skilled in the art can also obtain other drawings based on these drawings.

[0073] Figure 1 A schematic diagram of the structure of the performance monitoring system based on the embodiment of the present application;

[0074] Figure 2 A schematic diagram of a performance monitoring method according to an embodiment of the present invention;

[0075] Figure 3 A schematic diagram of the structure of a performance monitoring device provided in an embodiment of the present application;

[0076] Figure 4 A schematic diagram of the structure of another performance monitoring device provided in an embodiment of the present application;

[0077] Figure 5A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0078] The above drawings have shown clear embodiments of the present application, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0080] In addition, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. In the description of the following embodiments, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0081] In related technologies, with the rapid development of information technology, storage systems are increasingly widely used in various industries. In order to ensure the stability and performance of storage systems, it is particularly important to timely discover and handle performance problems. However, traditional performance monitoring and alarm mechanisms often lack flexibility and intelligence, resulting in frequent false alarms, which brings additional burdens to system administrators and may affect the overall performance of the system. Traditional alarm mechanisms are usually based on preset fixed thresholds, and once a performance indicator exceeds the threshold, an alarm is triggered. This method does not take into account factors such as hardware specification differences and business model changes in the storage system, and it is easy to mistake normal performance fluctuations for abnormalities, thereby generating false alarms. In addition, with the expansion of storage system scale and the increase in complexity, it becomes increasingly difficult to manually set and maintain thresholds, and an intelligent management method is urgently needed. Currently, there is no independent module in the industry to uniformly manage this intelligent performance monitoring and alarm mechanism. In a changing storage business environment, it is easy to cause cluster abnormal alarms or even fail to respond to front-end business normally, thereby affecting the availability of the entire system.

[0082] In response to the above problems, the performance monitoring method, device, electronic device and storage medium provided in the embodiments of the present application obtain the historical performance data of the storage system and the configuration change information of the storage system; for any performance indicator in the historical performance data, determine the performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system; determine the dynamic threshold corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period; determine the performance monitoring result of the storage system according to the current performance indicator of the storage system and the corresponding dynamic threshold. The method provided by the above scheme dynamically sets the performance indicator threshold by adapting to the configuration change information of the storage system, thereby avoiding mistaking normal performance fluctuations for abnormalities due to changes in the storage system configuration, thereby avoiding false alarms and improving the accuracy of the performance monitoring results.

[0083] The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0084] First, the structure of the performance monitoring system on which this application is based is described:

[0085] The performance monitoring method, device, electronic device, and storage medium provided in the embodiments of the present application are suitable for monitoring the performance of a storage system with a changed configuration. Figure 1 As shown, it is a schematic diagram of the structure of the performance monitoring system based on the embodiment of the present application, which mainly includes a data acquisition device, a storage system and a performance monitoring device. Among them, the data acquisition device is used to collect historical performance data of the storage system and configuration change information of the storage system, and send the collected data to the performance monitoring device, and the performance monitoring device dynamically determines the performance indicator threshold according to the obtained information, and then monitors the performance of the storage system according to the dynamic threshold to obtain the performance monitoring result.

[0086] The embodiment of the present application provides a performance monitoring method for monitoring the performance of a storage system with a changed configuration. The execution subject of the embodiment of the present application is an electronic device, such as a server, a desktop computer, a laptop computer, a tablet computer, and other electronic devices that can be used to monitor the performance of a storage system.

[0087] like Figure 2 FIG. 1 is a flow chart of a performance monitoring method provided in an embodiment of the present application, the method comprising:

[0088] Step 201: Acquire historical performance data of a storage system and configuration change information of the storage system.

[0089] Specifically, when a configuration change occurs in the storage system, the step of acquiring historical performance data of the storage system and configuration change information of the storage system is triggered.

[0090] Specifically, various performance indicator data can be obtained in real time through the storage system interface (such as API interface). The collected data is preprocessed, including data cleaning and data normalization, and then statistical analysis and big data processing technology are used to extract the performance characteristics of the storage system. For example, by analyzing data such as IOPS, read and write latency, the performance of the storage system under different business models (IO size, read and write, random / sequential, etc.) can be obtained to obtain the historical performance data of the storage system.

[0091] Among them, the historical performance data at least includes IOPS, bandwidth, read and write latency, and utilization of resources such as CPU, and the configuration change information of the storage system at least includes hardware configuration change information, business scenario change information, and business load pressure change information.

[0092] Step 202: for any performance indicator in the historical performance data, determine a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system.

[0093] Specifically, for any performance indicator, the impact of the system configuration change information on the performance indicator can be analyzed to determine the performance identification result corresponding to the performance indicator, and the performance identification result is used to characterize the normal fluctuation range of the performance indicator affected by the configuration change. Specifically, the collected data can be batch processed and analyzed using a big data processing platform such as Hadoop to extract the performance characteristics and fluctuation patterns of the storage system.

[0094] Step 203: Determine a dynamic threshold corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period.

[0095] Specifically, for any performance indicator, a dynamic threshold of the performance indicator can be determined based on performance identification results corresponding to multiple historical values ​​of the performance indicator within a preset period to ensure that the determined dynamic threshold adapts to the normal impact of system configuration changes on the performance indicator.

[0096] Step 204: Determine the performance monitoring result of the storage system according to the current performance index of the storage system and the corresponding dynamic threshold.

[0097] Specifically, by comparing the current performance indicator with the corresponding dynamic threshold, it can be determined whether the current performance indicator is a normal value, thereby obtaining the performance monitoring results of the storage system. The current performance indicator of the storage system can be displayed through a dashboard such as Grafana to display the performance status of the storage system in real time. When an abnormality is detected, the system automatically sends an email, SMS, system notification, etc. to notify the system administrator, and highlights the abnormal indicator on the dashboard.

[0098] Furthermore, the measures taken by the administrator can continue to be monitored and fed back to the machine learning module. When similar alarms occur in the future, possible measures can be output to the system administrator at the same time to help the system administrator resolve the alarm.

[0099] Specifically, in one embodiment, the current abnormal performance indicator can be determined based on the size relationship between the current performance indicator of the storage system and the corresponding dynamic threshold; performance alarm information is generated based on the current abnormal performance indicator to obtain the performance monitoring result of the storage system.

[0100] Specifically, based on the abnormal judgment criteria of various types of performance indicators, when the current performance indicator is greater than or less than the corresponding dynamic threshold, the performance indicator can be determined as the current abnormal performance indicator, and then corresponding performance alarm information can be generated to remind relevant technologies to perform technical maintenance in a timely manner according to the performance monitoring results of the storage system to improve the performance of the storage system.

[0101] On the basis of the above embodiments, since the determination of the dynamic threshold will directly affect the accuracy of the performance monitoring result of the storage system, in order to further improve the accuracy of the performance monitoring result, as an implementable method, in one embodiment, for any performance indicator in the historical performance data, according to the performance indicator and the configuration change information of the storage system, the performance identification result corresponding to the performance indicator is determined, including:

[0102] Step 2021, determining several change factors of the storage system according to the configuration change information of the storage system;

[0103] Step 222: for any performance indicator in the historical performance data, select a key change factor that affects the performance indicator from among several change factors;

[0104] Step 223, obtaining multiple performance identification models, where the performance identification models are constructed based on a machine learning algorithm and historical operation data of the storage system;

[0105] Step 2024: for any performance identification model, input the change degree of the key change factor and the performance index into the performance identification model, so as to identify the normal fluctuation of the performance index based on the performance identification model and the change degree of the key change factor and the performance index, and obtain the model output result;

[0106] Step 2025, assigning a weight coefficient to each performance identification model according to the confidence of the model output result of each performance identification model;

[0107] Step 2026: Determine the performance identification result corresponding to the performance indicator according to the model output results of each performance identification model and the weight coefficient of each performance identification model.

[0108] Among them, the change factors of the storage system include at least hardware changes, business changes and load pressure adjustments. Several performance identification models can be constructed based on decision trees, random forests, support vector machines, neural networks and other technologies.

[0109] Specifically, first, according to the configuration change information of the storage system, several types of change factors experienced by the storage system are comprehensively sorted out and determined. These change factors may include hardware configuration adjustment, software version update, system parameter modification and other aspects. Next, for each performance indicator in the historical performance data, the key change factors that affect the performance indicator are screened out from several change factors. This screening process requires in-depth analysis of the potential correlation between each change factor and the performance indicator to accurately identify the key factors that have a significant impact on the performance indicator. Then, multiple performance identification models are obtained. These performance identification models are based on advanced machine learning algorithms and are built using rich historical operation data of the storage system. Each model has been trained and has the ability to accurately identify and analyze the performance of the storage system. For any performance identification model, the degree of change of the key change factor and the performance indicator are input into the model. In this way, the model can accurately identify the normal fluctuation of the performance indicator based on its internal learning mechanism and algorithm according to the degree of change of the key change factor and the actual performance of the performance indicator, and output the corresponding model output result. In order to comprehensively consider the reliability and accuracy of each performance identification model, a corresponding weight coefficient is assigned to each model according to the confidence of the model output results of each performance identification model. Models with higher confidence will be given greater weights to reflect their importance and reliability in performance identification. Finally, based on the model output results of each performance identification model and the weight coefficients of each performance identification model, the final performance identification results corresponding to the performance indicators are determined through comprehensive calculation and analysis. This result will comprehensively consider the opinions and weights of multiple models, thereby providing a more comprehensive, accurate and reliable performance evaluation to obtain accurate performance identification results corresponding to each performance indicator.

[0110] Specifically, in one embodiment, the performance identification result corresponding to the performance indicator may be determined based on the following formula:

[0111]

[0112] Among them, f(x) represents the performance identification result corresponding to the performance indicator x, n represents the number of performance identification models, and w i represents the weight coefficient of performance identification model i, h i (x) represents the model output result of performance identification model i for performance indicator x, and b represents the variation factor bias.

[0113] Specifically, when the performance identification models constructed using different technologies obtain the same input performance index x, their output results h i (x) Since there are certain differences in the recognition advantages of different models, the embodiment of the present application improves the accuracy of the final performance recognition results by weighted fusion of the output results of all performance recognition models.

[0114] Furthermore, in one embodiment, for any performance indicator, the average performance identification result and the standard deviation of the performance identification result of the performance indicator can be determined based on the performance identification result corresponding to the performance indicator within a preset period; and the dynamic threshold corresponding to the performance indicator can be determined based on the average performance identification result and the standard deviation of the performance identification result of the performance indicator.

[0115] The dynamic threshold corresponding to the performance indicator is determined based on the following formula:

[0116] T=u+k·σ

[0117] Wherein, T represents the dynamic threshold corresponding to the performance indicator, u represents the average performance identification result of the performance indicator, k represents the adjustment coefficient, and σ represents the standard deviation of the performance identification result of the performance indicator.

[0118] Specifically, by calculating the dynamic threshold based on the above formula, it is ensured that the dynamic threshold can reflect the actual business situation and actual configuration of the storage system, etc., so as to reduce false alarms. For example, in a storage system, differences in hardware specifications may cause slight fluctuations in performance, but this does not mean that there are performance problems in the system. The embodiment of the present application avoids false alarms by identifying this situation and incorporating it into the normal performance fluctuation range. That is, when the hardware of the storage system is replaced or the business type changes, the alarm threshold is adjusted in time to ensure that the alarm mechanism can adapt to the new system state.

[0119] On the basis of the above embodiment, in order to further improve the accuracy of the performance monitoring result, as an implementable manner, in one embodiment, the method further includes:

[0120] Step 301, obtaining performance false alarm information fed back by users of the storage system;

[0121] Step 302: Optimize the dynamic threshold corresponding to each performance indicator according to the performance false alarm information.

[0122] Specifically, based on the false performance alarm information fed back by users, the causes of false alarms can be analyzed, and then the dynamic threshold settings can be adjusted based on the analysis, the machine learning model can be improved, and the correlation coefficients of the above formulas can be optimized, so as to continuously optimize the alarm strategy, reduce false alarms and missed alarms, and thus improve the accuracy of alarms.

[0123] Specifically, in one embodiment, in order to further improve the security of the storage system, the time series data of each performance indicator can also be recorded to obtain the historical time series data of each performance indicator. Then, based on the changing characteristics and prediction requirements of each performance indicator and the historical time series data of each performance indicator, the predicted value of each performance indicator at a certain point in the future is predicted. Based on the predicted value of the performance indicator and the dynamic threshold determined based on the above embodiment, the performance warning result of the storage system is determined to predict the performance anomaly of the storage system in advance.

[0124] Furthermore, the performance bottlenecks in the storage system can be determined based on the predicted values ​​and dynamic thresholds of the performance indicators. These bottlenecks may be caused by hardware limitations, improper configuration, business load, etc. If the performance bottleneck is caused by hardware limitations, you can consider upgrading the hardware equipment, such as increasing storage capacity, increasing disk read and write speeds, and increasing cache. You can also check the configuration parameters of the storage system, such as disk array settings, cache policies, file system parameters, etc., and optimize and adjust them according to performance requirements. You can also analyze the access pattern and distribution of data and optimize the distribution of data in the storage system to improve data access efficiency. You can also evaluate the impact of business processes on the storage system and find possible optimization points, such as reducing unnecessary data access and optimizing data backup policies. If there are multiple nodes or disks in the storage system, you can consider implementing a load balancing strategy to ensure that the load is evenly distributed and improve the overall performance of the system.

[0125] The performance monitoring method provided by the embodiment of the present application obtains the historical performance data of the storage system and the configuration change information of the storage system; for any performance indicator in the historical performance data, according to the performance indicator and the configuration change information of the storage system, determines the performance identification result corresponding to the performance indicator; according to the performance identification result corresponding to each performance indicator of the storage system within a preset period, determines the dynamic threshold corresponding to each performance indicator; according to the current performance indicator of the storage system and the corresponding dynamic threshold, determines the performance monitoring result of the storage system. The method provided by the above scheme dynamically sets the performance indicator threshold by adapting to the configuration change information of the storage system, avoids mistaking normal performance fluctuations for abnormalities due to changes in the storage system configuration, and thus avoids causing false alarms, thereby improving the accuracy of the performance monitoring results. In addition, through the management of centralized storage performance thresholds and alarm strategies based on machine learning and data analysis technology, false alarms can be effectively reduced, the accuracy and response speed of alarms can be improved, and the stability and efficient operation of the storage system can be ensured. Through intelligent management methods, it is possible to automatically identify and adapt to changes in the storage system, improve self-management capabilities, reduce manual intervention, and ultimately achieve comprehensive monitoring and optimization of storage system performance, thereby improving the availability of the overall system.

[0126] An embodiment of the present application provides a performance monitoring device for executing the performance monitoring method provided in the above embodiment.

[0127] like Figure 3 , which is a schematic diagram of the structure of a performance monitoring device provided in an embodiment of the present application. The performance monitoring device 30 comprises: an acquisition module 301 , a first determination module 302 , a second determination module 303 and a monitoring module 304 .

[0128] Among them, the acquisition module is used to obtain the historical performance data of the storage system and the configuration change information of the storage system; the first determination module is used to determine the performance identification result corresponding to any performance indicator in the historical performance data according to the performance indicator and the configuration change information of the storage system; the second determination module is used to determine the dynamic threshold corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period; the monitoring module is used to determine the performance monitoring result of the storage system according to the current performance indicator of the storage system and the corresponding dynamic threshold.

[0129] Specifically, in one embodiment, the first determining module is specifically configured to:

[0130] Determine several change factors of the storage system according to the configuration change information of the storage system;

[0131] For any performance indicator in the historical performance data, select the key change factors that affect the performance indicator from among several change factors;

[0132] Obtain multiple performance identification models, which are built based on machine learning algorithms and historical operation data of the storage system;

[0133] For any performance identification model, the degree of change of the key change factor and the performance index are input into the performance identification model, so as to identify the normal fluctuation of the performance index based on the performance identification model and the degree of change of the key change factor and the performance index, and obtain the model output result;

[0134] Allocating a weight coefficient to each performance identification model according to the confidence of the model output result of each performance identification model;

[0135] Determine the performance identification result corresponding to the performance indicator according to the model output results of each performance identification model and the weight coefficient of each performance identification model;

[0136] Among them, the change factors of the storage system include at least hardware change, business change and load pressure adjustment.

[0137] Specifically, in one embodiment, the first determining module is specifically configured to:

[0138] The performance identification result corresponding to the performance indicator is determined based on the following formula:

[0139]

[0140] Among them, f(x) represents the performance identification result corresponding to the performance indicator x, n represents the number of performance identification models, and w i represents the weight coefficient of performance identification model i, h i (x) represents the model output result of performance identification model i for performance indicator x, and b represents the variation factor bias.

[0141] Specifically, in one embodiment, the second determining module is specifically configured to:

[0142] For any performance indicator, according to the performance identification result corresponding to the performance indicator within a preset period, determine the average performance identification result and the standard deviation of the performance identification result of the performance indicator;

[0143] A dynamic threshold corresponding to the performance indicator is determined according to an average performance identification result and a standard deviation of the performance identification result of the performance indicator.

[0144] Specifically, in one embodiment, the second determining module is specifically configured to:

[0145] Based on the following formula, the dynamic threshold corresponding to the performance indicator is determined according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator:

[0146] T=u+k·σ

[0147] Wherein, T represents the dynamic threshold corresponding to the performance indicator, u represents the average performance identification result of the performance indicator, k represents the adjustment coefficient, and σ represents the standard deviation of the performance identification result of the performance indicator.

[0148] Specifically, in one embodiment, the monitoring module is specifically used to:

[0149] Determine the current abnormal performance indicator based on the size relationship between the current performance indicator of the storage system and the corresponding dynamic threshold;

[0150] Generate performance alarm information based on current abnormal performance indicators to obtain performance monitoring results of the storage system.

[0151] Specifically, in one embodiment, the device further includes:

[0152] The optimization module is used to obtain performance false alarm information fed back by users of the storage system; based on the performance false alarm information, the dynamic threshold corresponding to each performance indicator is optimized.

[0153] For example, as another division of logical functions, Figure 4 As shown, it is a structural schematic diagram of another performance monitoring device provided by an embodiment of the present application, and the device includes an intelligent performance monitoring management module (IPM management module), a data acquisition module, a data analysis module, a machine learning module, a threshold management module, an alarm strategy module, a monitoring and alarm module, and a parameter module. The IPM management module is located on the board, and mainly applies programmable logic devices such as ARM. The IPM management module can dynamically manage the data acquisition module, the data analysis module, the machine learning module, the threshold management module, the alarm strategy module, the monitoring and alarm module. The data acquisition module is responsible for collecting various types of performance data from the storage system, including but not limited to IOPS, bandwidth, read and write latency, CPU and other storage utilization. The data analysis module uses big data analysis technology to process and analyze the collected data to extract the performance characteristics and fluctuation patterns of the storage system. The machine learning module trains and updates the machine learning model based on historical performance data and the current business model to identify normal performance fluctuations and abnormal situations. The threshold management module dynamically sets and adjusts the threshold of performance monitoring according to the output of the machine learning module to ensure that the threshold can reflect the actual business situation of the storage system. The alarm strategy module designs and implements intelligent alarm strategies based on the threshold settings provided by the threshold management module, reduces false alarms, and can adjust in real time according to changes in the storage system. The monitoring and alarm module monitors the performance of the storage system in real time. When an abnormal situation is detected, an alarm is triggered and the system administrator is notified to take appropriate measures. The parameter module can be used to exchange information between the outside world and the IPM management module, preset parameters, and enable related functions.

[0154] Regarding the performance monitoring device in this embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0155] The performance monitoring device provided in the embodiment of the present application is used to execute the performance monitoring method provided in the above embodiment. Its implementation method and principle are the same and will not be repeated here.

[0156] An embodiment of the present application provides an electronic device for executing the performance monitoring method provided in the above embodiment.

[0157] like Figure 5 FIG. 5 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 50 includes: at least one processor 51 and a memory 52.

[0158] The memory stores computer-executable instructions; at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the performance monitoring method provided in the above embodiment.

[0159] The electronic device provided in the embodiment of the present application is used to execute the performance monitoring method provided in the above embodiment. Its implementation method and principle are the same and will not be repeated here.

[0160] An embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, a performance monitoring method provided in any of the above embodiments is implemented.

[0161] The storage medium containing computer executable instructions provided in the embodiment of the present application can be used to store the computer executable instructions of the performance monitoring method provided in the aforementioned embodiment. The implementation method and principle are the same and will not be repeated here.

[0162] An embodiment of the present application provides a computer program product, including computer instructions, which are used to enable a computer to execute the performance monitoring method provided in the aforementioned embodiment.

[0163] The embodiments of the present application provide a computer program product that can be used to execute computer instructions of the performance monitoring method provided in the aforementioned embodiments. The implementation method and principle are the same and will not be repeated here.

[0164] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0165] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0166] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0167] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0168] A part of the present application may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0169] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A performance monitoring method, characterized in that: include: Acquire historical performance data of a storage system and configuration change information of the storage system; For any performance indicator in the historical performance data, determine a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system; Determine a dynamic threshold value corresponding to each performance indicator according to the performance identification result corresponding to each performance indicator of the storage system within a preset period; A performance monitoring result of the storage system is determined according to a current performance indicator of the storage system and a corresponding dynamic threshold.

2. The method according to claim 1, characterized in that The step of determining, for any performance indicator in the historical performance data, a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system, includes: Determining several change factors of the storage system according to the configuration change information of the storage system; For any performance indicator in the historical performance data, selecting a key change factor that affects the performance indicator from among the several change factors; Acquire a plurality of performance identification models, wherein the performance identification models are constructed based on a machine learning algorithm and historical operation data of the storage system; For any of the performance identification models, the degree of change of the key change factor and the performance index are input into the performance identification model, so as to identify the normal fluctuation of the performance index based on the performance identification model and the degree of change of the key change factor and the performance index, and obtain a model output result; Allocating a weight coefficient to each of the performance identification models according to the confidence of the model output result of each of the performance identification models; Determining a performance identification result corresponding to the performance indicator according to the model output results of each of the performance identification models and the weight coefficients of each of the performance identification models; The change factors of the storage system include at least hardware change, service change and load pressure adjustment.

3. The method according to claim 2, characterized in that Determining the performance identification result corresponding to the performance indicator according to the model output results of each of the performance identification models and the weight coefficients of each of the performance identification models includes: The performance identification result corresponding to the performance indicator is determined based on the following formula: Among them, f(x) represents the performance identification result corresponding to the performance indicator x, n represents the number of performance identification models, and w i represents the weight coefficient of performance identification model i, h i (x) represents the model output result of performance identification model i for performance indicator x, and b represents the variation factor bias.

4. The method according to claim 1, characterized in that: The determining, according to the performance identification results corresponding to the performance indicators of the storage system within a preset period, the dynamic threshold corresponding to each performance indicator includes: For any of the performance indicators, according to the performance identification results corresponding to the performance indicator within a preset period, determine the average performance identification result and the standard deviation of the performance identification result of the performance indicator; A dynamic threshold corresponding to the performance indicator is determined according to an average performance identification result and a standard deviation of the performance identification result of the performance indicator.

5. The method according to claim 4, characterized in that The determining, according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator, the dynamic threshold value corresponding to the performance indicator comprises: Based on the following formula, the dynamic threshold corresponding to the performance indicator is determined according to the average performance identification result and the standard deviation of the performance identification result of the performance indicator: T=u+k·σ Wherein, T represents the dynamic threshold corresponding to the performance indicator, u represents the average performance identification result of the performance indicator, k represents the adjustment coefficient, and σ represents the standard deviation of the performance identification result of the performance indicator.

6. The method according to claim 1, characterized in that Determining the performance monitoring result of the storage system according to the current performance indicator of the storage system and the corresponding dynamic threshold includes: Determine the current abnormal performance indicator according to the magnitude relationship between the current performance indicator of the storage system and the corresponding dynamic threshold; According to the current abnormal performance indicator, performance alarm information is generated to obtain the performance monitoring result of the storage system.

7. The method according to claim 1, characterized in that The method further comprises: Obtain performance false alarm information from storage system users; According to the performance false alarm information, the dynamic threshold corresponding to each performance indicator is optimized.

8. A performance monitoring device, characterized in that: include: An acquisition module, used to acquire historical performance data of a storage system and configuration change information of the storage system; A first determination module is used to determine, for any performance indicator in the historical performance data, a performance identification result corresponding to the performance indicator according to the performance indicator and the configuration change information of the storage system; A second determination module, configured to determine a dynamic threshold corresponding to each performance indicator according to a performance identification result corresponding to each performance indicator of the storage system within a preset period; The monitoring module is used to determine the performance monitoring result of the storage system according to the current performance index of the storage system and the corresponding dynamic threshold.

9. An electronic device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when a processor executes the computer-executable instructions, the method according to any one of claims 1 to 7 is implemented.

Images