Performance-based system architecture rationality evaluation method and system

Abstract

The present invention discloses a performance-based system architecture rationality evaluation method and system, which belongs to the technical field of system architecture rationality evaluation. The system of the present invention includes: a gradient pressure test module, a data association and analysis module, a key point extraction module and a strategy adjustment and evaluation module; the gradient pressure test module executes a preset strategy, obtains the historical test data and performance indicators of the system, and saves them; the data association and analysis module matches the data according to the time mark, forms an associated data group and analyzes the association relationship between the historical test data and the historical performance indicators; the key point extraction module determines the key points of the gradient pressure test according to the association relationship; the strategy adjustment and evaluation module adjusts the strategy based on the key points of the gradient pressure test, uses the adjusted strategy to perform the gradient pressure test, analyzes the deviation relationship between the historical performance indicators and the real-time performance indicators, obtains the rationality evaluation coefficient, and evaluates the effectiveness of the adjusted gradient pressure strategy in combination with the deviation relationship.

Description

Technical Field

[0001] The present invention relates to the technical field of system architecture rationality evaluation, and in particular to a performance-based system architecture rationality evaluation method and system. Background Art

[0002] With the continuous advancement of technology and the growing business needs, the complexity of system architecture is also increasing. In order to ensure the rationality of the system architecture, relevant personnel often use performance testing to evaluate the performance of the system under various working conditions. Among them, the use of gradient pressure to obtain test data can reveal the performance bottlenecks of the system as a whole and each component at different pressure levels.

[0003] However, the current performance-based system architecture rationality assessment method and system still have some shortcomings. Although the existing technology has adopted the gradient pressure method, the gradient pressure strategy setting of some systems is not fine enough, resulting in inaccurate extraction of key points of gradient pressure test. The key points of gradient pressure test refer to the performance bottleneck points corresponding to the test task, and the rationality of the system architecture can be evaluated according to the performance bottleneck points; the existing technology is not fine enough for the gradient pressure strategy setting, resulting in inaccurate extraction of key points of gradient pressure test, which may lead to inaccurate test data and fail to truly reflect the performance of the system under different pressure levels, thus affecting the evaluation results of the rationality of the system architecture. Summary of the invention

[0004] The purpose of the present invention is to provide a performance-based system architecture rationality assessment method and system to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A performance-based system architecture rationality assessment method, the method comprising the following steps:

[0007] Step S100. Perform a gradient pressurization test on the system architecture to be evaluated according to a preset gradient pressurization strategy, thereby obtaining historical test data of the gradient pressurization test of the system architecture to be evaluated; based on the historical test data, obtain historical performance indicators of the system architecture to be evaluated, and both the historical test data and the historical performance indicators have time tags;

[0008] Step S200. Based on the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data are matched with the historical performance indicators to form a plurality of associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators;

[0009] Step S300. According to the correlation between the historical test data and the historical performance indicators, the key points of the gradient pressurization test of the system architecture to be evaluated are obtained; based on the key points of the gradient pressurization test, the preset gradient pressurization strategy is adjusted, and the adjusted gradient pressurization strategy is saved;

[0010] Step S400. Perform a gradient pressure test on the system architecture to be evaluated using the adjusted gradient pressure test strategy, and obtain real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyze the historical performance indicators and the real-time performance indicators to obtain the deviation relationship between the two;

[0011] Step S500. Based on the historical performance indicators and the real-time performance indicators, the rationality evaluation coefficients of the corresponding system architecture to be evaluated are obtained respectively, and the effectiveness of the adjusted gradient pressurization strategy is judged in combination with the deviation relationship between the historical performance indicators and the real-time performance indicators.

[0012] Furthermore, step S100 includes:

[0013] S101. The preset gradient pressure test strategy refers to the test strategy formulated by relevant personnel before the gradient pressure test according to the characteristics of the system architecture to be evaluated and the test objectives, and the formulated test strategy contains different test conditions, including the type of load to be applied, load intensity, duration and other key parameters; the gradient pressure test is performed on the system architecture to be evaluated according to the preset gradient pressure test strategy, so as to obtain the historical test data of the system architecture to be evaluated under different test conditions; the historical test data refers to the test data recorded and collected during the execution of the gradient pressure test; these data include performance indicators such as the system's response time, throughput, CPU and memory usage, as well as status information under different test conditions during system operation;

[0014] S102. According to the preset gradient pressurization strategy, obtain the test task time period corresponding to the preset gradient pressurization strategy, divide the historical test data according to the test task time period, divide it into several historical test data segments, and one historical test data segment corresponds to one test task, and one historical test data segment corresponds to one time mark, and the time mark refers to the time period mark between the start time point and the end time point of the corresponding test task; organize and analyze each historical test data segment to obtain the corresponding historical performance indicator, and the time mark of the historical performance indicator is consistent with the time mark of the corresponding historical test data segment.

[0015] Further, step S200 includes:

[0016] S201. According to the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data and the historical performance indicators with the same time stamp are matched to form a number of associated data groups, and the number of associated data groups is equal to the number of test tasks; for each associated data group, in chronological order, the historical test data and the historical performance indicators are respectively visualized using line graphs to obtain a historical test data line graph and a historical performance indicator line graph, and the number of the historical test data line graph and the historical performance indicator line graph is equal to the type of the corresponding data type;

[0017] S202. For all the data points of the historical test data line graph and the historical performance indicator line graph of each associated data group, they are numbered in chronological order, and the slopes k of adjacent numbered data points are calculated to form a slope set Ai and a slope set Bj, wherein the slope set Ai represents the slope set of the i-th historical test data line graph, the slope set Bj represents the slope set of the j-th historical performance indicator line graph, i represents the number of historical test data line graphs, j represents the number of historical performance indicator line graphs, and i and j are positive integers respectively; based on the slope set Ai and the slope set Bj, the correlation between the two is calculated respectively, and the average value is obtained to obtain the average correlation coefficient P; if the average correlation coefficient P is not equal to 0, it indicates that there is a linear relationship between the historical test data and the historical performance indicators; if the average correlation coefficient P is equal to 0, it indicates that there is no linear relationship between the historical test data and the historical performance indicators.

[0018] Furthermore, step S300 includes:

[0019] S301. Obtain the correlation analysis results of the historical test data and historical performance indicators of all associated data groups, respectively count the number of associated data groups M1 with linear relationship between historical test data and historical performance indicators and the number of associated data groups M2 with no linear relationship, and compare the size relationship between M1 and M2;

[0020] If M1 is greater than or equal to M2, then extract the associated data group in which the historical test data and the historical performance indicators have a linear relationship as the sample data group, perform linear regression analysis on each sample data group based on the corresponding historical test data and historical performance indicators, so as to obtain the corresponding linear regression equation Y1, perform verification analysis on the historical test data and historical performance indicators in the sample data group according to the linear regression equation Y1, and take the continuous time period corresponding to the historical test data and historical performance indicators that satisfy the linear regression equation Y1 as a cycle;

[0021] S302. If there is only one cycle in the sample data set and there is no data outside this cycle, the last time point corresponding to the above cycle is used as the key point of the gradient pressurization test corresponding to the current sample data set;

[0022] If there is only one cycle in the sample data group and there are historical test data and historical performance indicators outside this cycle, then the historical test data and historical performance indicators outside this cycle are obtained, and the historical test data and historical performance indicators outside this cycle are represented in a line graph. When the line graphs corresponding to the historical test data and historical performance indicators are both line segments parallel to the horizontal axis, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is used as the key point of the gradient pressurization test corresponding to the current sample data group; when the line graph corresponding to the historical test data and historical performance indicators does not have a line segment parallel to the horizontal axis, the last time point outside this cycle is used as the key point of the gradient pressurization test corresponding to the current sample data group;

[0023] If there are at least two cycles in the sample data group, then the time points corresponding to the start and end points of the first two adjacent cycles are obtained in chronological order. If the end time point of the previous cycle is different from the start time point of the next cycle, then the time point corresponding to the end time point of the previous cycle and the start time point of the next cycle and the corresponding historical test data and historical performance indicator line graph are obtained. If there is only one time point corresponding to the end time point of the previous cycle and the start time point of the next cycle, or the corresponding historical test data and historical performance indicator line graph do not have a line segment parallel to the horizontal axis, then , then the last time point between the end time point of the previous cycle and the time point corresponding to the starting point of the next cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group; if there is a line segment parallel to the horizontal axis in both the corresponding historical test data and the historical performance index line graph, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is taken as the key point of the gradient pressure test corresponding to the current sample data group; if the end time point of the previous cycle is the same as the starting time point of the next cycle, then the end time point of the previous cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group;

[0024] S303. Extract the associated data group with no linear relationship between the historical test data and the historical performance indicators as the data group to be adjusted, sort the sample data group and the data group to be adjusted in chronological order, obtain the linear regression equation Y1 of the closest sample data group, adjust the linear regression equation Y1 of the closest sample data group according to the historical test data and historical performance indicators corresponding to the data group to be adjusted, and use the adjusted linear regression equation Y1 of the closest sample data group as the linear regression equation Y2 corresponding to the data group to be adjusted, and if the historical test data and historical performance indicators corresponding to the data group to be adjusted both satisfy the linear regression equation Y2, then obtain the time period corresponding to the data group to be adjusted, and use the last time point of the time period corresponding to the data group to be adjusted as the tentative key point of the gradient pressurization test of the data group to be adjusted.

[0025] Furthermore, step S300 includes:

[0026] S304. If M1 is less than M2, obtain the historical test data line graphs and historical performance index line graphs corresponding to all associated data groups and the corresponding slope sets Ai and Bj; extract periodic sequences from the elements in the slope sets Ai and Bj, and if the time points corresponding to the periodic sequences in the slope sets Ai and Bj are exactly the same, obtain the number N1 and N2 of periodic sequences in the slope sets Ai and Bj of the current associated data group;

[0027] S305. If N1=N2=1, determine whether there are other elements in the slope set Ai and the slope set Bj of the current associated data group except the periodic sequence. If there are other elements, determine whether the other elements are equal to 0. If the other elements in the slope set Ai and the slope set Bj are all equal to 0, then the time point corresponding to the first element equal to 0 in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group; if there are other non-zero elements in the other elements in the slope set Ai and the slope set Bj, then the last identical time point in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group;

[0028] S306. If N1≠N2, then obtain the last identical time point in the slope set Ai and the slope set Bj of the current associated data group as the key point of the gradient pressurization test of the current associated data group; if N1=N2>1, then obtain the elements corresponding to the end point of the first period sequence and the starting point of the second period sequence in the slope set Ai and the slope set Bj of the current associated data group respectively, and if the above elements are all equal to zero, then obtain the first identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group; if the above elements are not all equal to zero, then obtain the last identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group;

[0029] S307. Obtain the key points of the gradient pressurization test, and adjust the corresponding preset gradient pressurization strategy according to the length between the key points of the gradient pressurization test and the initial points of the corresponding test tasks, so as to obtain the adjusted gradient pressurization strategy.

[0030] Furthermore, step S400 includes:

[0031] The gradient stress test is performed on the system architecture to be evaluated using the adjusted gradient stress test strategy, and the real-time test data and real-time performance indicators of the gradient stress test of the system architecture to be evaluated are obtained; the real-time test data and real-time performance indicators are divided according to the corresponding test tasks, so as to obtain the real-time test data segments that are the same as the historical test data and historical performance indicators; for each real-time test data segment, the historical test data segments corresponding to the same test task and the historical performance indicators and real-time performance indicators corresponding to the real-time test data segments are analyzed, and the deviation value MD between the historical performance indicators and the real-time performance indicators is calculated, and the specific formula is:

[0032]

[0033] Where n represents the number of data points, and the number of data points corresponding to the historical performance indicators and the real-time performance indicators is the same; h s Represents the value of the sth data point in the historical performance indicator; r s Represents the value of the sth data point in the real-time performance indicator.

[0034] Furthermore, step S500 includes:

[0035] S501. For each detection task, the rationality evaluation coefficient of the corresponding system architecture to be evaluated is obtained according to the historical performance indicators, which is recorded as the historical rationality evaluation coefficient G1; the rationality evaluation coefficient of the corresponding system architecture to be evaluated is obtained in the same way as the historical performance indicators, which is recorded as the real-time rationality evaluation coefficient G2; wherein, according to the existing system architecture rationality evaluation method or software, the corresponding performance indicators can be input to obtain the corresponding rationality evaluation coefficient; the deviation value MD1 between the historical rationality evaluation coefficient G1 and the real-time rationality evaluation coefficient G2 is calculated, and the specific calculation formula is: MD1 = |G1-G2| / |G1+G2|;

[0036] S502. Calculate the ratio Q between MD1 and MD, and compare the ratio Q with the corresponding deviation threshold Q0. If the ratio Q is less than or equal to the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be valid, and the adjusted gradient pressurization strategy corresponding to the current detection task is saved; if the ratio Q is greater than the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be invalid, then the adjusted gradient pressurization strategy corresponding to the current detection task is excluded, and the information that the adjusted gradient pressurization strategy corresponding to the current detection task is invalid is output to relevant personnel, and the relevant personnel perform corresponding processing.

[0037] A performance-based system architecture rationality evaluation system, which includes: a gradient pressure test module, a data association and analysis module, a key point extraction module, and a strategy adjustment and evaluation module;

[0038] The gradient pressure test module executes the preset gradient pressure strategy, obtains the historical test data and historical performance indicators of the system architecture to be evaluated, and saves the historical test data and historical performance indicators;

[0039] The data association and analysis module matches the historical test data with the historical performance indicators based on the time tags corresponding to the historical test data and the historical performance indicators, thereby forming several associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators;

[0040] The key point extraction module obtains the key points of the gradient stress test of the system architecture to be evaluated based on the correlation between the historical test data and the historical performance indicators;

[0041] The strategy adjustment and evaluation module adjusts the preset gradient pressure strategy based on the key points of the gradient pressure test, and saves the adjusted gradient pressure strategy; uses the adjusted gradient pressure strategy to perform gradient pressure test on the system architecture to be evaluated, and obtains real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyzes historical performance indicators and real-time performance indicators to obtain the deviation relationship between the two; obtains the rationality evaluation coefficient of the corresponding system architecture to be evaluated based on the historical performance indicators and the real-time performance indicators, and judges the effectiveness of the adjusted gradient pressure strategy based on the deviation relationship between the historical performance indicators and the real-time performance indicators.

[0042] Furthermore, the data association and analysis module includes a data matching unit, a line graph visualization unit, and a linear relationship judgment unit;

[0043] The data matching unit matches the data with the same time tags according to the time tags corresponding to the historical test data and the historical performance indicators to form an associated data group; the line chart visualization unit visualizes the historical test data and historical performance indicators of each associated data group; the linear relationship judgment unit judges whether there is a linear relationship between the historical test data and the historical performance indicators based on the slope set of the historical test data and the historical performance indicators.

[0044] Furthermore, the strategy adjustment and evaluation module includes a real-time data acquisition and analysis unit, a performance indicator deviation calculation unit, a rationality evaluation coefficient calculation unit, and an analysis and decision-making unit;

[0045] The real-time data acquisition and analysis unit acquires the real-time test data and real-time performance indicators of the system architecture to be evaluated, divides the real-time data into corresponding real-time test data segments according to the test task, and corresponds them with the historical test data segments of the same test task; the performance indicator deviation calculation unit compares and analyzes the historical performance indicator segment with the real-time performance indicator segment for each real-time test data segment, and calculates the deviation value between the historical performance indicator and the real-time performance indicator; the rationality evaluation coefficient calculation unit obtains the corresponding rationality evaluation coefficient for each detection task based on the historical performance indicator and the real-time performance indicator; the analysis and decision unit calculates the deviation value between the historical rationality evaluation coefficient and the real-time rationality evaluation coefficient, and judges the effectiveness of the adjusted gradient pressurization strategy corresponding to the current detection task in combination with the deviation value between the historical performance indicator and the real-time performance indicator, and outputs the corresponding judgment result.

[0046] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0047] By analyzing the correlation between historical test data and performance indicators, the key points of the gradient pressure test can be identified and adjusted. This ensures that during the test, the pressure changes of the system can more accurately simulate the pressure changes in actual business scenarios, avoid excessive or small gradient changes, and improve the accuracy and authenticity of the test; through linear regression analysis, the linear relationship between historical test data and performance indicators can be identified, and the data groups with linear relationships can be accurately found to ensure the accuracy and reliability of the key points of the gradient pressure test; if there is no linear relationship, the situation with multiple cycles can be analyzed, and the starting and end points of each cycle can be extracted in chronological order, and the selection of key points of the gradient pressure test can be further refined to reflect the changes and stability of system performance. The key points of the gradient pressurization test are not only the nodes of pressure change, but also the linear relationship of historical data, so as to more accurately reflect the performance of the system under different pressure levels; according to the key points obtained by analysis, the preset gradient pressurization strategy is automatically adjusted. This method can adjust the system architecture more timely to optimize the accuracy and real-time performance of performance evaluation; real-time performance testing is carried out under the adjusted gradient pressurization strategy, and the real-time performance indicators are compared and analyzed with the historical performance indicators. By analyzing the deviation relationship between the two, the effectiveness of the adjusted gradient pressurization strategy can be timely evaluated and verified to ensure the stability and performance of the system under different pressure levels, which is helpful to more comprehensively evaluate the advantages and disadvantages of the system architecture and the room for improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1It is a schematic diagram of the system modules of the performance-based system architecture rationality evaluation system of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 , the present invention provides a technical solution:

[0052] A performance-based system architecture rationality evaluation system, which includes: a gradient pressure test module, a data association and analysis module, a key point extraction module, and a strategy adjustment and evaluation module;

[0053] The gradient pressure test module executes the preset gradient pressure strategy, obtains the historical test data and historical performance indicators of the system architecture to be evaluated, and saves the historical test data and historical performance indicators;

[0054] The data association and analysis module matches the historical test data with the historical performance indicators based on the time tags corresponding to the historical test data and the historical performance indicators, thereby forming several associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators;

[0055] The key point extraction module obtains the key points of the gradient stress test of the system architecture to be evaluated based on the correlation between the historical test data and the historical performance indicators;

[0056] The strategy adjustment and evaluation module adjusts the preset gradient pressure strategy based on the key points of the gradient pressure test, and saves the adjusted gradient pressure strategy; uses the adjusted gradient pressure strategy to perform gradient pressure test on the system architecture to be evaluated, and obtains real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyzes historical performance indicators and real-time performance indicators to obtain the deviation relationship between the two; obtains the rationality evaluation coefficient of the corresponding system architecture to be evaluated based on the historical performance indicators and the real-time performance indicators, and judges the effectiveness of the adjusted gradient pressure strategy based on the deviation relationship between the historical performance indicators and the real-time performance indicators.

[0057] The data association and analysis module includes a data matching unit, a line graph visualization unit, and a linear relationship judgment unit;

[0058] The data matching unit matches the data with the same time tags according to the time tags corresponding to the historical test data and the historical performance indicators to form an associated data group; the line chart visualization unit visualizes the historical test data and historical performance indicators of each associated data group; the linear relationship judgment unit judges whether there is a linear relationship between the historical test data and the historical performance indicators based on the slope set of the historical test data and the historical performance indicators.

[0059] The strategy adjustment and evaluation module includes a real-time data acquisition and analysis unit, a performance indicator deviation calculation unit, a rationality evaluation coefficient calculation unit, and an analysis and decision-making unit;

[0060] The real-time data acquisition and analysis unit acquires the real-time test data and real-time performance indicators of the system architecture to be evaluated, divides the real-time data into corresponding real-time test data segments according to the test task, and corresponds them with the historical test data segments of the same test task; the performance indicator deviation calculation unit compares and analyzes the historical performance indicator segment with the real-time performance indicator segment for each real-time test data segment, and calculates the deviation value between the historical performance indicator and the real-time performance indicator; the rationality evaluation coefficient calculation unit obtains the corresponding rationality evaluation coefficient for each detection task based on the historical performance indicator and the real-time performance indicator; the analysis and decision unit calculates the deviation value between the historical rationality evaluation coefficient and the real-time rationality evaluation coefficient, and judges the effectiveness of the adjusted gradient pressurization strategy corresponding to the current detection task in combination with the deviation value between the historical performance indicator and the real-time performance indicator, and outputs the corresponding judgment result.

[0061] A performance-based system architecture rationality assessment method, the method comprising the following steps:

[0062] Step S100. Perform a gradient pressurization test on the system architecture to be evaluated according to a preset gradient pressurization strategy, thereby obtaining historical test data of the gradient pressurization test of the system architecture to be evaluated; based on the historical test data, obtain historical performance indicators of the system architecture to be evaluated, and both the historical test data and the historical performance indicators have time tags;

[0063] Step S200. Based on the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data are matched with the historical performance indicators to form a plurality of associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators;

[0064] Step S300. According to the correlation between the historical test data and the historical performance indicators, the key points of the gradient pressurization test of the system architecture to be evaluated are obtained; based on the key points of the gradient pressurization test, the preset gradient pressurization strategy is adjusted, and the adjusted gradient pressurization strategy is saved;

[0065] Step S400. Perform a gradient pressure test on the system architecture to be evaluated using the adjusted gradient pressure test strategy, and obtain real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyze the historical performance indicators and the real-time performance indicators to obtain the deviation relationship between the two;

[0066] Step S500. Based on the historical performance indicators and the real-time performance indicators, the rationality evaluation coefficients of the corresponding system architecture to be evaluated are obtained respectively, and the effectiveness of the adjusted gradient pressurization strategy is judged in combination with the deviation relationship between the historical performance indicators and the real-time performance indicators.

[0067] Step S100 includes:

[0068] S101. The preset gradient pressure test strategy refers to the test strategy formulated by relevant personnel before the gradient pressure test according to the characteristics of the system architecture to be evaluated and the test objectives, and the formulated test strategy contains different test conditions, including the type of load to be applied, load intensity, duration and other key parameters; the gradient pressure test is performed on the system architecture to be evaluated according to the preset gradient pressure test strategy, so as to obtain the historical test data of the system architecture to be evaluated under different test conditions; the historical test data refers to the test data recorded and collected during the execution of the gradient pressure test; these data include performance indicators such as the system's response time, throughput, CPU and memory usage, as well as status information under different test conditions during system operation;

[0069] S102. According to the preset gradient pressurization strategy, obtain the test task time period corresponding to the preset gradient pressurization strategy, divide the historical test data according to the test task time period, divide it into several historical test data segments, and one historical test data segment corresponds to one test task, and one historical test data segment corresponds to one time mark, and the time mark refers to the time period mark between the start time point and the end time point of the corresponding test task; organize and analyze each historical test data segment to obtain the corresponding historical performance indicator, and the time mark of the historical performance indicator is consistent with the time mark of the corresponding historical test data segment.

[0070] Step S200 includes:

[0071] S201. According to the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data and the historical performance indicators with the same time stamp are matched to form a number of associated data groups, and the number of associated data groups is equal to the number of test tasks; for each associated data group, in chronological order, the historical test data and the historical performance indicators are respectively visualized using line graphs to obtain a historical test data line graph and a historical performance indicator line graph, and the number of the historical test data line graph and the historical performance indicator line graph is equal to the type of the corresponding data type;

[0072] S202. For all the data points of the historical test data line graph and the historical performance indicator line graph of each associated data group, they are numbered in chronological order, and the slopes k of adjacent numbered data points are calculated to form a slope set Ai and a slope set Bj, wherein the slope set Ai represents the slope set of the i-th historical test data line graph, the slope set Bj represents the slope set of the j-th historical performance indicator line graph, i represents the number of historical test data line graphs, j represents the number of historical performance indicator line graphs, and i and j are positive integers respectively; based on the slope set Ai and the slope set Bj, the correlation between the two is calculated respectively, and the average value is obtained to obtain the average correlation coefficient P; if the average correlation coefficient P is not equal to 0, it indicates that there is a linear relationship between the historical test data and the historical performance indicators; if the average correlation coefficient P is equal to 0, it indicates that there is no linear relationship between the historical test data and the historical performance indicators.

[0073] Step S300 includes:

[0074] S301. Obtain the correlation analysis results of the historical test data and historical performance indicators of all associated data groups, respectively count the number of associated data groups M1 with linear relationship between historical test data and historical performance indicators and the number of associated data groups M2 with no linear relationship, and compare the size relationship between M1 and M2;

[0075] If M1 is greater than or equal to M2, then extract the associated data group in which the historical test data and the historical performance indicators have a linear relationship as the sample data group, perform linear regression analysis on each sample data group based on the corresponding historical test data and historical performance indicators, so as to obtain the corresponding linear regression equation Y1, perform verification analysis on the historical test data and historical performance indicators in the sample data group according to the linear regression equation Y1, and take the continuous time period corresponding to the historical test data and historical performance indicators that satisfy the linear regression equation Y1 as a cycle;

[0076] S302. If there is only one cycle in the sample data set and there is no data outside this cycle, the last time point corresponding to the above cycle is used as the key point of the gradient pressurization test corresponding to the current sample data set;

[0077] If there is only one cycle in the sample data group and there are historical test data and historical performance indicators outside this cycle, then the historical test data and historical performance indicators outside this cycle are obtained, and the historical test data and historical performance indicators outside this cycle are represented in a line graph. When the line graphs corresponding to the historical test data and historical performance indicators are both line segments parallel to the horizontal axis, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is used as the key point of the gradient pressurization test corresponding to the current sample data group; when the line graph corresponding to the historical test data and historical performance indicators does not have a line segment parallel to the horizontal axis, the last time point outside this cycle is used as the key point of the gradient pressurization test corresponding to the current sample data group;

[0078] If there are at least two cycles in the sample data group, then the time points corresponding to the start and end points of the first two adjacent cycles are obtained in chronological order. If the end time point of the previous cycle is different from the start time point of the next cycle, then the time point corresponding to the end time point of the previous cycle and the start time point of the next cycle and the corresponding historical test data and historical performance indicator line graph are obtained. If there is only one time point corresponding to the end time point of the previous cycle and the start time point of the next cycle, or the corresponding historical test data and historical performance indicator line graph do not have a line segment parallel to the horizontal axis, then , then the last time point between the end time point of the previous cycle and the time point corresponding to the starting point of the next cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group; if there is a line segment parallel to the horizontal axis in both the corresponding historical test data and the historical performance index line graph, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is taken as the key point of the gradient pressure test corresponding to the current sample data group; if the end time point of the previous cycle is the same as the starting time point of the next cycle, then the end time point of the previous cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group;

[0079] S303. Extract the associated data group with no linear relationship between the historical test data and the historical performance indicators as the data group to be adjusted, sort the sample data group and the data group to be adjusted in chronological order, obtain the linear regression equation Y1 of the closest sample data group, adjust the linear regression equation Y1 of the closest sample data group according to the historical test data and historical performance indicators corresponding to the data group to be adjusted, and use the adjusted linear regression equation Y1 of the closest sample data group as the linear regression equation Y2 corresponding to the data group to be adjusted, and if the historical test data and historical performance indicators corresponding to the data group to be adjusted both satisfy the linear regression equation Y2, then obtain the time period corresponding to the data group to be adjusted, and use the last time point of the time period corresponding to the data group to be adjusted as the tentative key point of the gradient pressurization test of the data group to be adjusted.

[0080] In this embodiment, assuming that a slope set Ai of an associated data group is {a1, a2, a3, a4, a5, a1, a2, a3, a4}, it can be obtained from the elements in the above set that there is periodicity in the slope set Ai, and the elements corresponding to one period are (a1, a2, a3, a4), which means that the slope set Ai can be decomposed into a specific periodic sequence of (a1, a2, a3, a4) that repeats; the time point corresponding to a5 is obtained as the provisional gradient pressurization test key point of the current associated data group; if there is periodicity in the slope set Bj of the current associated data group, a provisional gradient pressurization test key point is also obtained according to the analysis method of the slope set Ai, and if the time points corresponding to the two provisional gradient pressurization test key points are consistent, the above provisional gradient pressurization test key point is used as the gradient pressurization test key point of the current associated data group.

[0081] Step S300 includes:

[0082] S304. If M1 is less than M2, obtain the historical test data line graphs and historical performance index line graphs corresponding to all associated data groups and the corresponding slope sets Ai and Bj; extract periodic sequences from the elements in the slope sets Ai and Bj, and if the time points corresponding to the periodic sequences in the slope sets Ai and Bj are exactly the same, obtain the number N1 and N2 of periodic sequences in the slope sets Ai and Bj of the current associated data group;

[0083] S305. If N1=N2=1, determine whether there are other elements in the slope set Ai and the slope set Bj of the current associated data group except the periodic sequence. If there are other elements, determine whether the other elements are equal to 0. If the other elements in the slope set Ai and the slope set Bj are all equal to 0, then the time point corresponding to the first element equal to 0 in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group; if there are other non-zero elements in the other elements in the slope set Ai and the slope set Bj, then the last identical time point in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group;

[0084] S306. If N1≠N2, then obtain the last identical time point in the slope set Ai and the slope set Bj of the current associated data group as the key point of the gradient pressurization test of the current associated data group; if N1=N2>1, then obtain the elements corresponding to the end point of the first period sequence and the starting point of the second period sequence in the slope set Ai and the slope set Bj of the current associated data group respectively, and if the above elements are all equal to zero, then obtain the first identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group; if the above elements are not all equal to zero, then obtain the last identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group;

[0085] S307. Obtain the key points of the gradient pressurization test, and adjust the corresponding preset gradient pressurization strategy according to the length between the key points of the gradient pressurization test and the initial points of the corresponding test tasks, so as to obtain the adjusted gradient pressurization strategy.

[0086] Step S400 includes:

[0087] The gradient stress test is performed on the system architecture to be evaluated using the adjusted gradient stress test strategy, and the real-time test data and real-time performance indicators of the gradient stress test of the system architecture to be evaluated are obtained; the real-time test data and real-time performance indicators are divided according to the corresponding test tasks, so as to obtain the real-time test data segments that are the same as the historical test data and historical performance indicators; for each real-time test data segment, the historical test data segments corresponding to the same test task and the historical performance indicators and real-time performance indicators corresponding to the real-time test data segments are analyzed, and the deviation value MD between the historical performance indicators and the real-time performance indicators is calculated, and the specific formula is:

[0088]

[0089] Where n represents the number of data points, and the number of data points corresponding to the historical performance indicators and the real-time performance indicators is the same; h s Represents the value of the sth data point in the historical performance indicator; r s Represents the value of the sth data point in the real-time performance indicator.

[0090] Step S500 includes:

[0091] S501. For each detection task, the rationality evaluation coefficient of the corresponding system architecture to be evaluated is obtained according to the historical performance indicators, which is recorded as the historical rationality evaluation coefficient G1; the rationality evaluation coefficient of the corresponding system architecture to be evaluated is obtained in the same way as the historical performance indicators, which is recorded as the real-time rationality evaluation coefficient G2; wherein, according to the existing system architecture rationality evaluation method or software, the corresponding performance indicators can be input to obtain the corresponding rationality evaluation coefficient; the deviation value MD1 between the historical rationality evaluation coefficient G1 and the real-time rationality evaluation coefficient G2 is calculated, and the specific calculation formula is: MD1 = |G1-G2| / |G1+G2|;

[0092] S502. Calculate the ratio Q between MD1 and MD, and compare the ratio Q with the corresponding deviation threshold Q0. If the ratio Q is less than or equal to the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be valid, and the adjusted gradient pressurization strategy corresponding to the current detection task is saved; if the ratio Q is greater than the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be invalid, then the adjusted gradient pressurization strategy corresponding to the current detection task is excluded, and the information that the adjusted gradient pressurization strategy corresponding to the current detection task is invalid is output to relevant personnel, and the relevant personnel perform corresponding processing.

[0093] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0094] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A performance-based system architecture rationality evaluation method, characterized by: The method comprises the following steps: Step S100. Perform a gradient pressurization test on the system architecture to be evaluated according to a preset gradient pressurization strategy, thereby obtaining historical test data of the gradient pressurization test of the system architecture to be evaluated; based on the historical test data, obtain historical performance indicators of the system architecture to be evaluated, and both the historical test data and the historical performance indicators have time tags; Step S200. Based on the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data are matched with the historical performance indicators to form a plurality of associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators; Step S300. According to the correlation between the historical test data and the historical performance indicators, the key points of the gradient pressurization test of the system architecture to be evaluated are obtained; based on the key points of the gradient pressurization test, the preset gradient pressurization strategy is adjusted, and the adjusted gradient pressurization strategy is saved; The step S300 includes: S301. Obtain the correlation analysis results of the historical test data and historical performance indicators of all associated data groups, respectively count the number of associated data groups M1 with linear relationship between historical test data and historical performance indicators and the number of associated data groups M2 with no linear relationship, and compare the size relationship between M1 and M2; If M1 is greater than or equal to M2, then extract the associated data group in which the historical test data and the historical performance indicators have a linear relationship as the sample data group, perform linear regression analysis on each sample data group based on the corresponding historical test data and historical performance indicators, so as to obtain the corresponding linear regression equation Y1, perform verification analysis on the historical test data and historical performance indicators in the sample data group according to the linear regression equation Y1, and take the continuous time period corresponding to the historical test data and historical performance indicators that satisfy the linear regression equation Y1 as a cycle; S302. If there is only one cycle in the sample data set and there is no data outside this cycle, the last time point corresponding to the above cycle is used as the key point of the gradient pressurization test corresponding to the current sample data set; If there is only one cycle in the sample data group and there are historical test data and historical performance indicators outside this cycle, then the historical test data and historical performance indicators outside this cycle are obtained, and the historical test data and historical performance indicators outside this cycle are represented in a line graph. When the line graphs corresponding to the historical test data and historical performance indicators are both line segments parallel to the horizontal axis, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is used as the key point of the gradient pressurization test corresponding to the current sample data group; when the line graph corresponding to the historical test data and historical performance indicators does not have a line segment parallel to the horizontal axis, the last time point outside this cycle is used as the key point of the gradient pressurization test corresponding to the current sample data group; If there are at least two cycles in the sample data group, then the time points corresponding to the start and end points of the first two adjacent cycles are obtained in chronological order. If the end time point of the previous cycle is different from the start time point of the next cycle, then the time point corresponding to the end time point of the previous cycle and the start time point of the next cycle and the corresponding historical test data and historical performance indicator line graph are obtained. If there is only one time point corresponding to the end time point of the previous cycle and the start time point of the next cycle, or the corresponding historical test data and historical performance indicator line graph do not have a line segment parallel to the horizontal axis, then , then the last time point between the end time point of the previous cycle and the time point corresponding to the starting point of the next cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group; if there is a line segment parallel to the horizontal axis in both the corresponding historical test data and the historical performance index line graph, the time point corresponding to the starting point of the line segment parallel to the horizontal axis is taken as the key point of the gradient pressure test corresponding to the current sample data group; if the end time point of the previous cycle is the same as the starting time point of the next cycle, then the end time point of the previous cycle is taken as the key point of the gradient pressure test corresponding to the current sample data group; S303. Extract the associated data group with no linear relationship between the historical test data and the historical performance indicators as the data group to be adjusted, sort the sample data group and the data group to be adjusted in chronological order, obtain the linear regression equation Y1 of the closest sample data group, adjust the linear regression equation Y1 of the closest sample data group according to the historical test data and historical performance indicators corresponding to the data group to be adjusted, and use the adjusted linear regression equation Y1 of the closest sample data group as the linear regression equation Y2 corresponding to the data group to be adjusted, and if the historical test data and historical performance indicators corresponding to the data group to be adjusted both satisfy the linear regression equation Y2, then obtain the time period corresponding to the data group to be adjusted, and use the last time point of the time period corresponding to the data group to be adjusted as the tentative key point of the gradient pressurization test of the data group to be adjusted; S304. If M1 is less than M2, obtain the historical test data line graphs and historical performance index line graphs corresponding to all associated data groups and the corresponding slope sets Ai and Bj; extract periodic sequences from the elements in the slope sets Ai and Bj, and if the time points corresponding to the periodic sequences in the slope sets Ai and Bj are exactly the same, obtain the number N1 and N2 of periodic sequences in the slope sets Ai and Bj of the current associated data group; S305. If N1=N2=1, determine whether there are other elements in the slope set Ai and the slope set Bj of the current associated data group except the periodic sequence. If there are other elements, determine whether the other elements are equal to 0. If the other elements in the slope set Ai and the slope set Bj are all equal to 0, then the time point corresponding to the first element equal to 0 in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group; if there are other non-zero elements in the other elements in the slope set Ai and the slope set Bj, then the last identical time point in the slope set Ai and the slope set Bj is used as the key point of the gradient pressurization test of the current associated data group; S306. If N1≠N2, then obtain the last identical time point in the slope set Ai and the slope set Bj of the current associated data group as the key point of the gradient pressurization test of the current associated data group; if N1=N2>1, then obtain the elements corresponding to the end point of the first period sequence and the starting point of the second period sequence in the slope set Ai and the slope set Bj of the current associated data group respectively, and if the above elements are all equal to zero, then obtain the first identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group; if the above elements are not all equal to zero, then obtain the last identical time point of the above elements in the slope set Ai and the slope set Bj as the key point of the gradient pressurization test of the current associated data group; S307. Obtaining the key points of the gradient pressurization test, and adjusting the corresponding preset gradient pressurization strategy according to the length between the key points of the gradient pressurization test and the initial points of the corresponding test tasks, thereby obtaining an adjusted gradient pressurization strategy; Step S400. Perform a gradient pressure test on the system architecture to be evaluated using the adjusted gradient pressure test strategy, and obtain real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyze the historical performance indicators and the real-time performance indicators to obtain the deviation relationship between the two; Step S500. Based on the historical performance indicators and the real-time performance indicators, the rationality evaluation coefficients of the corresponding system architecture to be evaluated are obtained respectively, and the effectiveness of the adjusted gradient pressurization strategy is judged in combination with the deviation relationship between the historical performance indicators and the real-time performance indicators.

2. The performance-based system architecture rationality evaluation method according to claim 1, characterized in that: The step S100 includes: S101. The preset gradient pressure test strategy refers to a test strategy formulated by relevant personnel according to the characteristics of the system architecture to be evaluated and the test objectives before the gradient pressure test, and the formulated test strategy includes different test conditions; the gradient pressure test is performed on the system architecture to be evaluated according to the preset gradient pressure test strategy, so as to obtain historical test data of the system architecture to be evaluated under different test conditions; the historical test data refers to the test data recorded and collected during the execution of the gradient pressure test; S102. According to the preset gradient pressurization strategy, obtain the test task time period corresponding to the preset gradient pressurization strategy, divide the historical test data according to the test task time period, divide it into several historical test data segments, and one historical test data segment corresponds to one test task, and one historical test data segment corresponds to one time mark, and the time mark refers to the time period mark between the start time point and the end time point of the corresponding test task; organize and analyze each historical test data segment to obtain the corresponding historical performance indicator, and the time mark of the historical performance indicator is consistent with the time mark of the corresponding historical test data segment.

3. The performance-based system architecture rationality evaluation method according to claim 2 is characterized in that: The step S200 includes: S201. According to the time stamps corresponding to the historical test data and the historical performance indicators, the historical test data and the historical performance indicators with the same time stamp are matched to form a number of associated data groups, and the number of associated data groups is equal to the number of test tasks; for each associated data group, in chronological order, the historical test data and the historical performance indicators are respectively visualized using line graphs to obtain a historical test data line graph and a historical performance indicator line graph, and the number of the historical test data line graph and the historical performance indicator line graph is equal to the type of the corresponding data type; S202. For all the data points of the historical test data line graph and the historical performance indicator line graph of each associated data group, they are numbered in chronological order, and the slopes k of adjacent numbered data points are calculated to form a slope set Ai and a slope set Bj, wherein the slope set Ai represents the slope set of the i-th historical test data line graph, the slope set Bj represents the slope set of the j-th historical performance indicator line graph, i represents the number of historical test data line graphs, j represents the number of historical performance indicator line graphs, and i and j are positive integers respectively; based on the slope set Ai and the slope set Bj, the correlation between the two is calculated respectively, and the average value is obtained to obtain the average correlation coefficient P; if the average correlation coefficient P is not equal to 0, it indicates that there is a linear relationship between the historical test data and the historical performance indicators; if the average correlation coefficient P is equal to 0, it indicates that there is no linear relationship between the historical test data and the historical performance indicators.

4. The performance-based system architecture rationality evaluation method according to claim 1, characterized in that: The step S400 includes: The gradient stress test is performed on the system architecture to be evaluated using the adjusted gradient stress test strategy, and the real-time test data and real-time performance indicators of the gradient stress test of the system architecture to be evaluated are obtained; the real-time test data and real-time performance indicators are divided according to the corresponding test tasks, so as to obtain the real-time test data segments that are the same as the historical test data and historical performance indicators; for each real-time test data segment, the historical test data segments corresponding to the same test task and the historical performance indicators and real-time performance indicators corresponding to the real-time test data segments are analyzed, and the deviation value MD between the historical performance indicators and the real-time performance indicators is calculated, and the specific formula is: Where n represents the number of data points, and the number of data points corresponding to the historical performance indicators and the real-time performance indicators is the same; h s Represents the value of the sth data point in the historical performance indicator; r s Represents the value of the sth data point in the real-time performance indicator.

5. The performance-based system architecture rationality evaluation method according to claim 4 is characterized in that: The step S500 includes: S501. For each detection task, obtain the rationality evaluation coefficient of the corresponding system architecture to be evaluated according to the historical performance indicators, recorded as the historical rationality evaluation coefficient G1; obtain the rationality evaluation coefficient of the corresponding system architecture to be evaluated in the same way as the historical performance indicators, recorded as the real-time rationality evaluation coefficient G2; calculate the deviation value MD1 between the historical rationality evaluation coefficient G1 and the real-time rationality evaluation coefficient G2, and the specific calculation formula is: MD1 = |G1-G2| / |G1+G2|; S502. Calculate the ratio Q between MD1 and MD, and compare the ratio Q with the corresponding deviation threshold Q0. If the ratio Q is less than or equal to the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be valid, and the adjusted gradient pressurization strategy corresponding to the current detection task is saved; if the ratio Q is greater than the deviation threshold Q0, then the adjusted gradient pressurization strategy corresponding to the current detection task is judged to be invalid, then the adjusted gradient pressurization strategy corresponding to the current detection task is excluded, and the information that the adjusted gradient pressurization strategy corresponding to the current detection task is invalid is output to relevant personnel, and the relevant personnel perform corresponding processing.

6. A system for evaluating the rationality of a performance-based system architecture, applied to a method for evaluating the rationality of a performance-based system architecture as claimed in any one of claims 1 to 5, characterized in that: The system includes: a gradient pressure test module, a data association and analysis module, a key point extraction module and a strategy adjustment and evaluation module; The gradient pressurization test module executes a preset gradient pressurization strategy, obtains historical test data and historical performance indicators of the system architecture to be evaluated, and saves the historical test data and historical performance indicators; The data association and analysis module matches the historical test data with the historical performance indicators based on the time tags corresponding to the historical test data and the historical performance indicators, thereby forming a number of associated data groups; each associated data group is analyzed to obtain the association relationship between the historical test data and the historical performance indicators; The key point extraction module obtains the key points of the gradient pressure test of the system architecture to be evaluated according to the correlation between the historical test data and the historical performance indicators; The strategy adjustment and evaluation module adjusts the preset gradient pressure strategy based on the key points of the gradient pressure test, and saves the adjusted gradient pressure strategy; uses the adjusted gradient pressure strategy to perform a gradient pressure test on the system architecture to be evaluated, and obtains real-time test data and real-time performance indicators of the gradient pressure test of the system architecture to be evaluated; analyzes the historical performance indicators and the real-time performance indicators to obtain the deviation relationship between the two; obtains the rationality evaluation coefficient of the corresponding system architecture to be evaluated based on the historical performance indicators and the real-time performance indicators, and judges the effectiveness of the adjusted gradient pressure strategy in combination with the deviation relationship between the historical performance indicators and the real-time performance indicators.

7. The performance-based system architecture rationality assessment system according to claim 6, characterized in that: The data association and analysis module includes a data matching unit, a line graph visualization unit, and a linear relationship judgment unit; The data matching unit matches the data with the same time tags according to the time tags corresponding to the historical test data and the historical performance indicators to form an associated data group; the line chart visualization unit visualizes the historical test data and the historical performance indicators of each associated data group; the linear relationship judgment unit judges whether there is a linear relationship between the historical test data and the historical performance indicators based on the slope set of the historical test data and the historical performance indicators.

8. The performance-based system architecture rationality assessment system according to claim 6, characterized in that: The strategy adjustment and evaluation module includes a real-time data acquisition and analysis unit, a performance indicator deviation calculation unit, a rationality evaluation coefficient calculation unit, and an analysis and decision-making unit; The real-time data acquisition and analysis unit acquires the real-time test data and real-time performance indicators of the system architecture to be evaluated, divides the real-time data into corresponding real-time test data segments according to the test task, and corresponds them with the historical test data segments of the same test task; the performance indicator deviation calculation unit compares and analyzes the historical performance indicator segment with the real-time performance indicator segment for each real-time test data segment, and calculates the deviation value between the historical performance indicator and the real-time performance indicator; the rationality evaluation coefficient calculation unit obtains the corresponding rationality evaluation coefficient for each detection task based on the historical performance indicator and the real-time performance indicator; the analysis and decision unit calculates the deviation value between the historical rationality evaluation coefficient and the real-time rationality evaluation coefficient, and judges the effectiveness of the adjusted gradient pressurization strategy corresponding to the current detection task in combination with the deviation value between the historical performance indicator and the real-time performance indicator, and outputs the corresponding judgment result.

Images