Computer performance prediction method, device, equipment and readable storage medium

Abstract

The present invention relates to the field of artificial intelligence technology, and discloses a computer performance prediction method, comprising the following steps: collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data; fitting the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model; adjusting the initial parameters of the initial timing analysis model to obtain a timing analysis model; predicting the computer performance data to be predicted using the timing analysis model to obtain a prediction result. The present invention also discloses a computer performance prediction device, equipment, and readable storage medium. The computer performance prediction method provided by the present invention improves the accuracy of computer performance prediction.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a computer performance prediction method, device, equipment and readable storage medium. Background Art

[0002] At present, predictions are often only made based on the historical data of the software, and the prediction conclusions are lagging, so potential problems cannot be discovered in time, and it is impossible to make correct assessments and warnings of future computer performance. Once the computer performance declines, it will take a lot of manpower and time to repair and improve it. At present, there is no better solution that can both predict the potential risks in computer performance and improve the prediction effect. How to improve the accuracy of computer performance prediction is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0003] The main purpose of the present invention is to provide a computer performance prediction method, device, equipment and readable storage medium, aiming to solve the technical problem of low accuracy of computer performance prediction.

[0004] To achieve the above object, the present invention provides a computer performance prediction method, which comprises the following steps:

[0005] Collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data;

[0006] Fitting the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T;

[0007] Adjusting initial parameters of the initial timing analysis model to obtain a target timing analysis model;

[0008] The target timing analysis model is used to predict the computer performance data to be predicted, and a prediction result is obtained.

[0009] Optionally, the computer performance data is fitted using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q and a period T, and the following steps are included:

[0010] Determining whether the computer performance data is non-stationary time series data;

[0011] If the computer performance data is non-stationary time series data, a d-order difference operation is performed to convert the non-stationary time series data into stationary time series data;

[0012] If the computer performance data is stationary time series data, then the autocorrelation coefficient and partial autocorrelation coefficient of the stationary time series data are calculated;

[0013] Based on the autocorrelation coefficient and the partial autocorrelation coefficient, the autocorrelation graph and the partial autocorrelation graph are analyzed to obtain the initial parameters of the initial timing analysis model in the preset timing analysis model set, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: differential coefficient d, autoregressive order p, sliding average order q and period T.

[0014] Optionally, the adjusting the initial parameters of the initial timing analysis model to obtain a target timing analysis model comprises the following steps:

[0015] Fitting the preset computer performance data through the initial timing analysis model in an exhaustive manner to obtain a first fitting result;

[0016] Calculating the goodness of fit of the first fitting result;

[0017] Determining whether the goodness of fit of the first fitting result is greater than or equal to a first preset threshold;

[0018] If the goodness of fit is greater than or equal to the first preset threshold, a target time series analysis model is obtained;

[0019] If the goodness of fit of the first fitting result is less than the first preset threshold, the initial parameters of the initial timing analysis model are adjusted until the goodness of fit is greater than or equal to the first preset threshold, thereby obtaining a target timing analysis model.

[0020] Optionally, predicting the computer performance data to be predicted by using the timing analysis model set to obtain a prediction result comprises the following steps:

[0021] Classifying the computer performance data into different types according to the frequency of occurrence of the computer performance data over time, wherein the different types include a periodic computer performance data class and / or a non-periodic computer performance data class;

[0022] According to the type of the computer performance data, a target type of a target timing analysis model is determined, and the computer performance data to be predicted is predicted by using the timing analysis model of the target type to obtain a prediction result.

[0023] Optionally, determining a target type of a target timing analysis model according to different types of the computer performance data, and predicting the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result comprises the following steps:

[0024] Determining whether the type of the computer performance data is a periodic computer performance data type;

[0025] If the type of the computer performance data is a periodic computer performance data type, the Prophet model is used to predict the computer performance data to obtain a first prediction result, and the first prediction result is used as the prediction result;

[0026] If the type of the computer performance data is a non-periodic computer performance data type, the ARIMA model is used to predict the computer performance data to obtain a second prediction result, and the second prediction result is used as the prediction result.

[0027] Optionally, after predicting the computer performance data to be predicted by the target timing analysis model to obtain a prediction result, the computer performance prediction method further includes the following steps:

[0028] Determining whether there are multiple first sub-prediction results or multiple second sub-prediction results;

[0029] If there are multiple first sub-prediction results or multiple second sub-prediction results, linear fitting is performed on the multiple first sub-prediction results or the multiple second sub-prediction results using the least squares method to obtain a second fitting result.

[0030] Optionally, if there are multiple first sub-prediction results or multiple second sub-prediction results, then after performing linear fitting on the multiple first sub-prediction results or the multiple second sub-prediction results by the least squares method to obtain a second fitting result, the computer performance prediction method further includes the following steps:

[0031] Calculating the goodness of fit of the second fitting result;

[0032] Determining whether the goodness of fit of the second fitting result is less than or equal to a second preset threshold;

[0033] If the goodness of fit of the second fitting result is less than or equal to a second preset threshold, the parameters of the target time series analysis model are adjusted until the goodness of fit of the second fitting result is greater than the second preset threshold.

[0034] Furthermore, to achieve the above object, the present invention also provides a computer performance prediction device, which includes the following modules:

[0035] A collection module, used for collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data;

[0036] A fitting module, used for fitting the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the initial timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T;

[0037] An adjustment module, used for adjusting initial parameters of the initial timing analysis model to obtain a target timing analysis model;

[0038] The prediction module is used to predict the computer performance data to be predicted through the target timing analysis model to obtain a prediction result.

[0039] Optionally, the fitting module includes the following units:

[0040] A judging unit, used to judge whether the computer performance data is non-stationary time series data;

[0041] A calculation unit, for performing a d-order difference calculation to convert the non-stationary time series data into stationary time series data if the computer performance data is non-stationary time series data, and for calculating an autocorrelation coefficient and a partial autocorrelation coefficient of the stationary time series data if the computer performance data is stationary time series data;

[0042] An analysis unit is used to analyze the autocorrelation diagram and the partial autocorrelation diagram based on the autocorrelation coefficient and the partial autocorrelation coefficient to obtain the initial parameters of the initial timing analysis model in the preset timing analysis model set, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: difference coefficient d, autoregressive order p, sliding average order q and period T.

[0043] Optionally, the adjustment module includes the following units:

[0044] A fitting unit, configured to fit the preset computer performance data through the initial timing analysis model in an exhaustive manner to obtain a first fitting result;

[0045] A goodness of fit calculation unit, used to calculate the goodness of fit of the first fitting result;

[0046] A goodness of fit judgment unit, used to judge whether the first fitting result is greater than or equal to a first preset threshold;

[0047] A timing analysis model acquisition unit, configured to obtain a target timing analysis model if the first fitting result is greater than or equal to the first preset threshold;

[0048] An initial parameter adjustment unit is used to adjust the initial parameters of the initial timing analysis model if the first fitting result is less than the first preset threshold, until the goodness of fit is greater than or equal to the first preset threshold, so as to obtain a target timing analysis model.

[0049] Optionally, the prediction module includes the following units:

[0050] a classification unit, for classifying the computer performance data into different types according to the frequency of occurrence of the computer performance data over time, wherein the different types include a periodic computer performance data class and / or a non-periodic computer performance data class;

[0051] The prediction unit is used to determine the target type of the target timing analysis model according to the type of the computer performance data, and predict the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result.

[0052] Optionally, the prediction unit is used to:

[0053] Determining whether the type of the computer performance data is a periodic computer performance data type;

[0054] If the type of the computer performance data is a periodic computer performance data type, the Prophet model is used to predict the computer performance data to obtain a first prediction result, and the first prediction result is used as the prediction result;

[0055] If the type of the computer performance data is a non-periodic computer performance data type, the ARIMA model is used to predict the computer performance data to obtain a second prediction result, and the second prediction result is used as the prediction result.

[0056] Optionally, the computer performance prediction device further includes the following modules:

[0057] A detection module, used to determine whether there are multiple first sub-prediction results or multiple second sub-prediction results;

[0058] A linear fitting module is used to perform linear fitting on the multiple first sub-prediction results or multiple second sub-prediction results by the least squares method to obtain a second fitting result if there are multiple first sub-prediction results or multiple second sub-prediction results.

[0059] Optionally, the computer performance prediction device further includes the following modules:

[0060] A second goodness of fit calculation module, used to calculate the goodness of fit of the second fitting result;

[0061] A second goodness of fit judgment module, used to judge whether the goodness of fit of the second fitting result is less than or equal to a second preset threshold;

[0062] An adjustment module is used to adjust the parameters of the target timing analysis model if the goodness of fit of the second fitting result is less than or equal to the second preset threshold value, until the goodness of fit of the second fitting result is greater than the second preset threshold value.

[0063] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a computer performance prediction device, which includes a memory, a processor, and a computer performance prediction program stored in the memory and executable on the processor, and the computer performance prediction program, when executed by the processor, implements the steps of the computer performance prediction method as described in any one of the above items.

[0064] Furthermore, to achieve the above-mentioned purpose, the present invention also provides a readable storage medium, on which a computer performance prediction program is stored, and when the computer performance prediction program is executed by a processor, the steps of the computer performance prediction method as described in any one of the above items are implemented.

[0065] The present invention predicts the computer performance data that may appear in the software development process through the target timing analysis model in the timing analysis model set, and can early detect whether the various performances of the computer are abnormal. The accuracy of the prediction results can be improved by processing the data with periodic changes and the data with non-periodic changes through different models. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A schematic diagram of the structure of the operating environment of a computer performance prediction device involved in an embodiment of the present invention;

[0067] Figure 2 It is a flowchart of a first embodiment of a method for predicting computer performance according to the present invention;

[0068] Figure 3 for Figure 2 A detailed flow chart of an embodiment of step S20;

[0069] Figure 4 for Figure 2 A detailed flow chart of an embodiment of step S30;

[0070] Figure 5 for Figure 2 A detailed flow chart of an embodiment of step S40;

[0071] Figure 6 for Figure 5 A detailed flow chart of an embodiment of step S402;

[0072] Figure 7 It is a flowchart of a second embodiment of the computer performance prediction method of the present invention;

[0073] Figure 8 Schematic diagram of the flow chart of the third embodiment of the computer performance prediction method of the present invention;

[0074] Fig. 9 The figure is a functional module diagram of an embodiment of a computer performance prediction device of the present invention.

[0075] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0076] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0077] The invention provides a computer performance prediction device.

[0078] Reference Figure 1 , Figure 1It is a structural schematic diagram of the operating environment of the computer performance prediction device involved in the embodiment of the present invention.

[0079] like Figure 1 As shown, the computer performance prediction device includes: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the network interface 1004 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface). The memory 1005 can be a high-speed RAM memory, or it can be a stable memory (non-volatile memory), such as a disk memory. The memory 1005 can also be a storage device independent of the aforementioned processor 1001.

[0080] Those skilled in the art will understand that Figure 1 The hardware structure of the computer performance prediction device shown in the figure does not constitute a limitation on the computer performance prediction device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0081] like Figure 1 As shown, the memory 1005 as a readable storage medium may include an operating system, a network communication module, a user interface module, and a computer performance prediction program. The operating system is a program that manages and controls computer performance prediction devices and software resources, and supports the operation of the computer performance prediction program and other software and / or programs.

[0082] exist Figure 1 In the hardware structure of the computer performance prediction device shown, the network interface 1004 is mainly used to access the network; the user interface 1003 is mainly used to detect confirmation instructions and editing instructions, etc. The processor 1001 can be used to call the computer performance prediction program stored in the memory 1005 and perform the operations of the following computer performance prediction method embodiments.

[0083] Based on the hardware structure of the computer performance prediction device described above, various embodiments of the computer performance prediction method of the present invention are proposed.

[0084] Reference Figure 2 , Figure 2 1 is a flow chart of a first embodiment of a computer performance prediction method of the present invention. In this embodiment, the computer performance prediction method comprises the following steps:

[0085] Step S10, collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data;

[0086] In this embodiment, the computer performance data may be CPU usage data and memory usage data. After the computer performance data are collected, they may be stored in chronological order or in iterative cycle order, thereby forming computer performance data sorted in chronological order.

[0087] Step S20, using an initial timing analysis model in a preset timing analysis model set to fit the computer performance data, and obtaining initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T;

[0088] In this embodiment, the preset timing analysis model set may include multiple timing analysis models, and different timing analysis models may fit different data. For example, if CPU usage data and memory usage data increase with time, the Prophet model may be used.

[0089] Step S30, adjusting the initial parameters of the initial timing analysis model to obtain a target timing analysis model;

[0090] In this embodiment, since the initial timing analysis model does not have the ability to accurately predict the computer performance data to be predicted, it is necessary to adjust the parameters to optimize the model.

[0091] Step S40, predicting the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result.

[0092] In this embodiment, because the computer performance data may contain both data that changes periodically and data that does not change periodically, after predicting different data, it is possible to obtain one prediction result or two prediction results. For example, the first prediction result is used to represent the prediction result obtained after predicting the data that changes periodically, and the second prediction result is used to represent the prediction result obtained after predicting the data that changes non-periodically.

[0093] By predicting quality problems that may occur during software development using the time series analysis model in the time series analysis model set, problems can be discovered early and early warnings can be provided for project management. The accuracy of prediction results can be improved by processing periodic and non-periodic data using different models.

[0094] Reference Figure 3 , Figure 3 for Figure 2 A detailed flow chart of an embodiment of step S20 in the embodiment. In this embodiment, step S20 uses an initial timing analysis model in a preset timing analysis model set to fit the computer performance data to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: differential coefficient d, autoregressive order p, sliding average order q and period T, including the following steps:

[0095] Step S201, determining whether the computer performance data is non-stationary time series data;

[0096] In this embodiment, time series data refers to data generated by a single variable in chronological order. For a time series, if the mean and variance do not have systematic or periodic changes, if the mean does not change, it means there is no obvious trend, and if the variance does not change, it means that the fluctuation is relatively stable, which is called stationary time series data. Non-stationary time series data does not have an unchanging central trend, and the distribution characteristics of the random variables of the data at each time point cannot be inferred by the sample mean and variance of the time series. Therefore, it can be judged whether the computer performance data is non-stationary time series data based on the above characteristics.

[0097] Step S202, if the computer performance data is non-stationary time series data, a d-order difference operation is performed to convert the non-stationary time series data into stationary time series data;

[0098] Step S203, if the computer performance data is stationary time series data, then the autocorrelation coefficient and partial autocorrelation coefficient of the stationary time series data are calculated;

[0099] In this embodiment, the d-order difference operation is a technical means to convert non-stationary time series data into stationary time series data. After the non-stationary time series data is converted into stationary time series data, the distribution characteristics of the random variables of the data at each time point can be easily obtained.

[0100] Calculate the autocorrelation coefficient ACF and partial autocorrelation coefficient PACF of stationary time series data. Both coefficients are used to represent the relationship between variables. When studying the influence of one variable on another variable, if the influence of other variables is considered at the same time, it is the correlation coefficient. If the influence of other variables is not considered, it is the partial autocorrelation coefficient. The autocorrelation coefficient is used to describe the degree of correlation between the data itself in different periods, that is, to measure the impact of historical data on current data.

[0101] Step S204, based on the autocorrelation coefficient and the partial autocorrelation coefficient, analyze the autocorrelation graph and the partial autocorrelation graph to obtain the initial parameters of the initial timing analysis model in the preset timing analysis model set, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: differential coefficient d, autoregressive order p, sliding average order q and period T.

[0102] In this embodiment, the initial parameters of the time series analysis model can be obtained by analyzing the autocorrelation graph and the partial autocorrelation graph. For example, when the order of the autocorrelation coefficient is 2nd or 3rd order, it quickly drops to near 0, that is, after removing the middle 2 or 3 variables, the sequence begins to stabilize, and the parameters at this time are taken as the initial parameters.

[0103] Reference Figure 4 , Figure 4 for Figure 2 A detailed flow chart of an embodiment of step S30 in FIG. In this embodiment, step S30, adjusting the initial parameters of the initial timing analysis model to obtain the target timing analysis model, includes the following steps:

[0104] Step S301, fitting preset computer performance data through an initial timing analysis model in an exhaustive manner to obtain a first fitting result;

[0105] In this embodiment, the computer performance data is fitted through the time series analysis model set in an exhaustive manner, so that the computer performance data in various scenarios can be covered. For example, if 900 data appear in a week, all of these data can be input into the time series analysis model set, and the time series analysis model set can be used to fit these data.

[0106] Step S302, calculating the goodness of fit of the first fitting result;

[0107] In this embodiment, the goodness of fit is the degree of fit of the regression line to the observed value, and the goodness of fit of the fitting result is calculated based on the preset fitting result.

[0108] Step S303, determining whether the goodness of fit of the first fitting result is greater than or equal to a first preset threshold;

[0109] In this embodiment, in order to obtain a time series analysis model set that satisfies the goodness of fit, it is necessary to determine whether the goodness of fit of the first fitting result is greater than or equal to a first preset threshold.

[0110] Step S304, if the goodness of fit of the first fitting result is greater than or equal to a first preset threshold, a target time series analysis model is obtained;

[0111] Step S305: if the first fitting result is less than the first preset threshold, the initial parameters of the initial time series analysis model are adjusted until the goodness of fit is greater than or equal to the first preset threshold, thereby obtaining a target time series analysis model.

[0112] In this embodiment, the basis for adjusting the initial parameters of the initial time series analysis model is to determine whether the probability of occurrence of each category in the population is consistent with the known probability. For example, to test whether a dice is even, the dice can be thrown several times, and the number of times each side appears can be recorded. Based on these data, it can be tested whether the probability of occurrence of each side is one-sixth. The goodness of fit test is used to test whether the distribution of the population of a batch of classified data is consistent with a certain theoretical distribution.

[0113] Reference Figure 5 , Figure 5 for Figure 2 A detailed flow chart of an embodiment of step S40 in the embodiment. In this embodiment, step S40, predicting the computer performance data to be predicted by the target timing analysis model to obtain a prediction result, includes the following steps:

[0114] Step S401, classifying the computer performance data into different types according to the frequency of occurrence of the computer performance data over time, wherein the different types include a periodic computer performance data class and / or a non-periodic computer performance data class;

[0115] In this embodiment, the computer performance data is divided into periodic computer performance data and non-periodic computer performance data in advance according to the frequency of occurrence of the computer performance data over time. If the data changes periodically over time, the data is set as periodic computer performance data, otherwise, it is set as non-periodic computer performance data.

[0116] Step S402, determining the target type of the target timing analysis model according to the type of the computer performance data, and predicting the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result.

[0117] In this embodiment, in order to obtain more accurate prediction results, it is necessary to select a suitable model for prediction according to the type of data. Prior to this, it is necessary to establish a mapping relationship between different types of computer performance data and different types of timing analysis models in advance, so as to determine the type of timing analysis model according to the type of computer performance data.

[0118] Reference Figure 6 , Figure 6 for Figure 5A detailed flow chart of an embodiment of step S402 in the embodiment. In this embodiment, step S402 determines the type of timing analysis model according to different types of computer performance data, and predicts the computer performance data to be predicted by the timing analysis model of the target type to obtain the prediction result, including the following steps:

[0119] Step S4021, determining whether the type of the computer performance data is a periodic computer performance data type;

[0120] Step S4022: if the type of the computer performance data is the periodic computer performance data type, the Prophet model is used to predict the computer performance data to obtain a first prediction result, and the first prediction result is used as the prediction result;

[0121] In this embodiment, since the Prophet model has a better prediction effect on data that changes periodically, the Prophet model is used for prediction. The Prophet prediction algorithm can establish a model based on historical data. The data of the periodic computer performance data class is input into the above model, and the prediction results within one month, one year or other periods are obtained according to the historical change trend of the data.

[0122] Step S4023, if the type of the computer performance data is a non-periodic computer performance data class, the ARIMA model is used to predict the computer performance data to obtain a second prediction result, and the second prediction result is used as the prediction result.

[0123] In this embodiment, since the ARIMA model has a better prediction effect on data with non-periodic changes, the ARIM model is used for prediction. The ARIMA model, or autoregressive integrated moving average model, is a relatively traditional prediction model that approximates the data series formed over time through the ARIMA model. When a prediction request is received, the model can predict future values ​​from the past and present values ​​of the time series.

[0124] Reference Figure 7 , Figure 7 1 is a flow chart of a second embodiment of the computer performance prediction method of the present invention. In this embodiment, in step S40, the computer performance data to be predicted is predicted by using the target timing analysis model, and after the prediction result is obtained, the computer performance prediction method further includes the following steps:

[0125] Step S50, determining whether there are multiple first sub-prediction results or multiple second sub-prediction results;

[0126] In this embodiment, since data can be predicted by multiple models, multiple prediction results can be obtained. There can be multiple models processing the same type of data, so multiple sub-prediction results can be obtained.

[0127] Step S60: If there are multiple first sub-prediction results or multiple second sub-prediction results, linear fitting is performed on the multiple first sub-prediction results or multiple second sub-prediction results by least square method to obtain a second fitting result.

[0128] In this embodiment, since multiple models can simultaneously predict the same type of data, in order to obtain the best prediction result, it is necessary to linearly fit multiple sub-prediction results. Since multiple sub-prediction results are obtained by predicting multiple models of the same type, there is also a connection between the prediction results of each model. The multiple sub-prediction results can be fitted by the least squares method to obtain a fitting result. The multiple sub-prediction results can be fitted into a curve by the least squares method. In this way, the change trend of the prediction result can be intuitively obtained.

[0129] Reference Figure 8 , Figure 8 1 is a flow chart of the third embodiment of the computer performance prediction method of the present invention. In this embodiment, in step S60, if there are multiple first sub-prediction results or multiple second sub-prediction results, the multiple first sub-prediction results or multiple second sub-prediction results are linearly fitted by the least square method to obtain the second fitting result, and the computer performance prediction method further includes the following steps:

[0130] Step S70, calculating the goodness of fit of the second fitting result;

[0131] In this embodiment, in order to obtain a time series analysis model set that satisfies the second goodness of fit, a first fitting result and a second fitting result are calculated.

[0132] Step S80, determining whether the goodness of fit of the second fitting result is less than or equal to a second preset threshold;

[0133] In this embodiment, it is determined whether the second goodness of fit is less than or equal to a preset second goodness of fit, and the second preset threshold is used as a reference for evaluating the second goodness of fit.

[0134] Step S90: If the goodness of fit of the second fitting result is less than or equal to the second preset threshold, the parameters of the target time series analysis model are adjusted until the goodness of fit of the second fitting result is greater than the second preset threshold.

[0135] In this embodiment, the preset timing analysis model set includes multiple timing analysis models, some of which are used to predict periodic computer performance data, and other timing analysis models are used to predict non-periodic computer performance data. Different models have different priorities. The higher the weight, the higher the priority. When there is an instruction to predict the computer performance data to be predicted, the timing analysis model with a high priority is used for prediction. Whether the second goodness of fit is less than or equal to the second preset threshold is judged as the standard for evaluating whether the model weight is appropriate. When it is less than or equal to the second preset threshold, it is necessary to adjust the weights of different models until the second goodness of fit is greater than the second preset threshold.

[0136] By predicting quality problems that may occur during software development using the timing analysis models in the preset timing analysis model set, problems can be discovered early and early warnings can be provided for project management. The accuracy of prediction results can be improved by processing periodic and non-periodic data using different models.

[0137] Reference Fig. 9 , Fig. 9 Schematic diagram of the functional modules of an embodiment of the computer performance prediction device of the present invention. In this embodiment, the computer performance prediction device includes:

[0138] A collecting module 10, for collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data;

[0139] A fitting module 20, used to fit the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T;

[0140] An adjustment module 30, used to adjust initial parameters of the initial timing analysis model to obtain a target timing analysis model;

[0141] The prediction module 40 is used to predict the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result.

[0142] In this embodiment, the modules of the device can process data with periodic changes and data with non-periodic changes respectively, which can improve the accuracy of the prediction results.

[0143] The invention also provides a readable storage medium.

[0144] In this embodiment, a computer performance prediction program is stored on the readable storage medium, and when the computer performance prediction program is executed by the processor, the steps of the computer performance prediction method described in any of the above embodiments are implemented.

[0145] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM) and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0146] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims. All equivalent structures or equivalent process changes made using the contents of the specification and drawings of the present invention, or directly or indirectly used in other related technical fields, are protected by the present invention.

Claims

1. A computer performance prediction method, characterized in that: The computer performance prediction method comprises the following steps: Collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data; Fitting the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T; Adjusting initial parameters of the initial timing analysis model to obtain a target timing analysis model; Predicting the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result; The adjusting the initial parameters of the initial timing analysis model to obtain the target timing analysis model includes the following steps: fitting the preset computer performance data through the initial timing analysis model in an exhaustive manner to obtain a first fitting result; calculating the goodness of fit of the first fitting result; judging whether the goodness of fit of the first fitting result is greater than or equal to a first preset threshold; if the goodness of fit of the first fitting result is greater than or equal to the first preset threshold, obtaining the target timing analysis model; if the first fitting result is less than the first preset threshold, adjusting the initial parameters of the initial timing analysis model until the goodness of fit is greater than or equal to the first preset threshold, thereby obtaining the target timing analysis model; The method of predicting the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result includes the following steps: dividing the computer performance data into different types according to the frequency of occurrence of the computer performance data over time, wherein the different types include periodic computer performance data classes and / or non-periodic computer performance data classes; determining the target type of the target timing analysis model according to the type of the computer performance data, and predicting the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result.

2. The computer performance prediction method according to claim 1, characterized in that: The computer performance data is fitted using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T, and the following steps are included: Determining whether the computer performance data is non-stationary time series data; If the computer performance data is non-stationary time series data, a d-order difference operation is performed to convert the non-stationary time series data into stationary time series data; If the computer performance data is stationary time series data, then the autocorrelation coefficient and partial autocorrelation coefficient of the stationary time series data are calculated; Based on the autocorrelation coefficient and the partial autocorrelation coefficient, the autocorrelation graph and the partial autocorrelation graph are analyzed to obtain the initial parameters of the initial timing analysis model in the preset timing analysis model set, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: differential coefficient d, autoregressive order p, sliding average order q and period T.

3. The computer performance prediction method according to claim 1, characterized in that: The method of determining the target type of the target timing analysis model according to the type of the computer performance data, and predicting the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result comprises the following steps: Determining whether the type of the computer performance data is a periodic computer performance data type; If the type of the computer performance data is a periodic computer performance data type, the Prophet model is used to predict the computer performance data to obtain a first prediction result, and the first prediction result is used as the prediction result; If the type of the computer performance data is a non-periodic computer performance data type, the ARIMA model is used to predict the computer performance data to obtain a second prediction result, and the second prediction result is used as the prediction result.

4. The computer performance prediction method according to claim 1, characterized in that: After predicting the computer performance data to be predicted by the target timing analysis model and obtaining the prediction result, the computer performance prediction method further includes the following steps: Determining whether there are multiple first sub-prediction results or multiple second sub-prediction results; If there are multiple first sub-prediction results or multiple second sub-prediction results, linear fitting is performed on the multiple first sub-prediction results or the multiple second sub-prediction results using the least squares method to obtain a second fitting result.

5. The computer performance prediction method according to claim 4, characterized in that: If there are multiple first sub-prediction results or multiple second sub-prediction results, then after performing linear fitting on the multiple first sub-prediction results or the multiple second sub-prediction results by the least square method to obtain a second fitting result, the computer performance prediction method further includes the following steps: Calculating the goodness of fit of the second fitting result; Determining whether the goodness of fit of the second fitting result is less than or equal to a second preset threshold; If the goodness of fit of the second fitting result is less than or equal to a second preset threshold, the parameters of the target time series analysis model are adjusted until the goodness of fit of the second fitting result is greater than the second preset threshold.

6. A computer performance prediction device, characterized in that: The computer performance prediction device comprises the following modules: A collection module, used for collecting computer performance data, wherein the computer performance data includes periodic computer performance data and non-periodic computer performance data; A fitting module, used to fit the computer performance data using an initial timing analysis model in a preset timing analysis model set to obtain initial parameters of the initial timing analysis model, wherein the preset timing analysis model set includes a first model for predicting periodic computer performance data and a second model for predicting non-periodic computer performance data, and the initial parameters include: a differential coefficient d, an autoregressive order p, a sliding average order q, and a period T; An adjustment module, used for adjusting initial parameters of the initial timing analysis model to obtain a target timing analysis model; A prediction module, used to predict the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result; The adjusting the initial parameters of the initial timing analysis model to obtain the target timing analysis model includes the following steps: fitting the preset computer performance data through the initial timing analysis model in an exhaustive manner to obtain a first fitting result; calculating the goodness of fit of the first fitting result; judging whether the goodness of fit of the first fitting result is greater than or equal to a first preset threshold; if the goodness of fit of the first fitting result is greater than or equal to the first preset threshold, obtaining the target timing analysis model; if the first fitting result is less than the first preset threshold, adjusting the initial parameters of the initial timing analysis model until the goodness of fit is greater than or equal to the first preset threshold, thereby obtaining the target timing analysis model; The method of predicting the computer performance data to be predicted by using the target timing analysis model to obtain a prediction result includes the following steps: dividing the computer performance data into different types according to the frequency of occurrence of the computer performance data over time, wherein the different types include periodic computer performance data classes and / or non-periodic computer performance data classes; determining the target type of the target timing analysis model according to the type of the computer performance data, and predicting the computer performance data to be predicted by using the timing analysis model of the target type to obtain a prediction result.

7. A computer performance prediction device, characterized in that: The computer performance prediction device includes a memory, a processor, and a computer performance prediction program stored in the memory and executable on the processor. When the computer performance prediction program is executed by the processor, the steps of the computer performance prediction method according to any one of claims 1 to 5 are implemented.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer performance prediction program, which, when executed by a processor, implements the steps of the computer performance prediction method according to any one of claims 1 to 5.

Images