Server case management method and system based on data monitoring

Abstract

The invention is suitable for the technical field of case data monitoring, and provides a server case management method and system based on data monitoring, and the method comprises the steps: determining a specific type of heavy load operation and a real-time noise average amplitude of a power supply path of a target unit at an initial stage when the target unit starts to execute the heavy load operation, and meanwhile, obtaining a preliminary risk prediction value and a historical execution record matched with the target unit. According to the method, the noise data of the power supply path of the target unit are collected in real time, the standard noise amplitude interval is established in combination with the historical execution record, and the noise interference caused by the parasitic parameter change of PCB wiring is accurately identified by using the dynamic spectrogram generated by the transient current fluctuation test; therefore, a correction factor is generated by comprehensively calculating the similar trend and the amplitude deviation, and the preliminary risk prediction value is dynamically corrected.

Description

Technical Field

[0001] The present invention belongs to the technical field of chassis data monitoring, and particularly relates to a server chassis management method and system based on data monitoring. Background Technique

[0002] At present, as an important part of data centers and cloud computing infrastructures, the stability and security of server chassis directly affect the overall system operation efficiency. Existing technologies mainly rely on monitoring conventional parameters such as current, voltage, and temperature to predict and manage risks of server chassis. However, in practical applications, these parameters often cannot fully reveal potential hazards caused by minute changes inside the electrical system. Especially during long-term use of PCB traces, their parasitic parameters will change, thereby causing abnormal fluctuations in the noise of the power supply path.

[0003] Such abnormal noise may hide potential electrical fault risks at the initial stage of high-load operations, and its change amplitude and dynamic characteristics are usually relatively weak. Traditional monitoring means are difficult to capture and correct this hidden danger in a timely manner, resulting in insufficient accuracy of risk assessment and affecting the overall effect of early warning and intelligent operation and maintenance.

[0004] In addition, existing management systems lack in-depth analysis means for noise characteristics in data processing and dynamic risk correction, resulting in delays and false alarms when detecting potential risks at the initial stage, thus restricting the improvement of the overall safety and stability of server chassis. Summary of the Invention

[0005] The purpose of the present invention is to provide a server chassis management method and system based on data monitoring, aiming to solve the problems raised in the background technique.

[0006] The present invention is implemented as follows. A server chassis management method based on data monitoring, the method includes:

[0007] At the starting stage when a target unit begins to execute a heavy-load operation, determine the specific type of the heavy-load operation and the real-time noise average amplitude of the power supply path of the target unit, and at the same time obtain a preliminary risk prediction value and a historical execution record matching the target unit;

[0008] Analyze the historical execution record, determine the standard noise amplitude range when the target unit executes a specific type of heavy-load operation, and judge whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset time period;

[0009] Obtain the test noise amplitude and the parasitic parameter variation value during each transient current fluctuation test, and accordingly draw a first dynamic frequency spectrum diagram and a second dynamic frequency spectrum diagram, and compare whether the proportion of similar trend points in the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram is greater than a preset threshold;

[0010] If it is determined that the proportion of similar trend points is greater than the preset threshold, then determine a set of key frequency points based on the first dynamic frequency spectrum diagram, and determine the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point;

[0011] According to the proportion of similar trend points and the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points, comprehensively correct the preliminary risk prediction value.

[0012] As a further limitation of the technical solution of the embodiment of the present invention, parse the historical execution record, determine the standard noise amplitude range when the target unit executes a specific type of heavy-load operation, and determine whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be yes, the steps of performing several transient current fluctuation tests on the PCB wiring of the target unit within a preset period include:

[0013] Parse the historical execution record, and screen out a predetermined number of secondary execution records that are consistent with the specific type and have no abnormalities during operation;

[0014] Analyze each secondary execution record, determine its corresponding standard noise amplitude, perform denoising and smoothing processing on the obtained standard noise amplitude, and generate a standard noise amplitude range with the maximum and minimum values among the processed several standard noise amplitudes as the upper and lower bounds respectively;

[0015] Determine whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset period.

[0016] As a further limitation of the technical solution of the embodiment of the present invention, the steps of obtaining the test noise amplitude and the parasitic parameter variation value during each transient current fluctuation test, and accordingly drawing a first dynamic frequency spectrum diagram and a second dynamic frequency spectrum diagram, and comparing whether the proportion of similar trend points in the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram is greater than a preset threshold include:

[0017] During each transient current fluctuation test, use a high-speed data acquisition device to collect the test noise amplitude, the parasitic parameter variation value, and the frequency range of each test in real time;

[0018] Construct a first dynamic frequency spectrum diagram based on the test noise amplitude and the frequency range, and construct a second dynamic frequency spectrum diagram based on the parasitic parameter variation value and the frequency range;

[0019] Compare and analyze the first dynamic spectrogram and the second dynamic spectrogram. By calculating the correlation coefficient of the signal amplitude changes between the two spectrograms in each frequency interval, when the correlation coefficient exceeds a preset value, the frequency interval is regarded as a similar trend point. Count the proportion of similar trend points in the total key frequency interval, and determine whether the proportion is greater than a preset threshold.

[0020] As a further limitation of the technical solution of the embodiment of the present invention, if it is determined that the proportion of similar trend points is greater than the preset threshold, the steps of determining the set of key frequency points based on the first dynamic spectrogram and determining the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point include:

[0021] Analyze the first dynamic spectrogram to determine a number of key frequency points and their corresponding actual noise amplitudes in the first dynamic spectrogram, and form a set of key frequency points by combining the number of key frequency points;

[0022] Analyze a predetermined number of secondary execution records, screen out a number of noise amplitude data consistent with each key frequency point, calculate the average value of the number of noise amplitudes corresponding to each key frequency point, and use it as the reference noise amplitude corresponding to each key frequency point.

[0023] As a further limitation of the technical solution of the embodiment of the present invention, the steps of comprehensively correcting the preliminary risk prediction value according to the proportion of similar trend points and the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points include:

[0024] Calculate and generate a first correction factor according to the proportion of similar trend points, and calculate and generate a second correction factor based on the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points;

[0025] Retrieve the preset correction formula, and apply the first correction factor and the second correction factor to the preliminary risk prediction value in combination to achieve the comprehensive correction of the preliminary risk prediction value.

[0026] As a further limitation of the technical solution of the embodiment of the present invention, the correction formula is: , where refers to the corrected risk prediction value, refers to the preliminary risk prediction value, refers to the first correction factor, refers to the second correction factor;

[0027] In the correction formula, , where refers to the calibration coefficient for adjusting the influence weight of the first correction factor, refers to the proportion of similar trend points;

[0028] , where refers to the calibration coefficient that adjusts the influence weight of the second correction factor, refers to the set of key frequency points, refers to belonging to the set of key frequency points a certain key frequency point in, refers to the key frequency point the corresponding actual noise amplitude, refers to the key frequency point the corresponding reference noise amplitude, refers to the key frequency point the corresponding weight.

[0029] A server chassis management system based on data monitoring, the system includes: a data acquisition module, a transient test module, a spectrogram drawing module, a set generation module, and a predicted value correction module, where:

[0030] The data acquisition module is used to determine the specific type of the heavy-load operation and the real-time noise average amplitude of the power supply path of the target unit at the start stage when the target unit starts to execute the heavy-load operation, and at the same time obtain the preliminary risk prediction value and historical execution record matching the target unit;

[0031] The transient test module is used to analyze the historical execution record, determine the standard noise amplitude range when the target unit executes a specific type of heavy-load operation, and judge whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be yes, several transient current fluctuation tests are performed on the PCB wiring of the target unit within a preset time period;

[0032] The spectrogram drawing module is used to obtain the test noise amplitude and parasitic parameter variation value during each transient current fluctuation test, and draw the first dynamic spectrogram and the second dynamic spectrogram based on this, and compare whether the proportion of similar trend points in the first dynamic spectrogram and the second dynamic spectrogram is greater than a preset threshold;

[0033] The set generation module is used to, if it is determined that the proportion of similar trend points is greater than the preset threshold, determine the set of key frequency points based on the first dynamic spectrogram, and determine the reference noise amplitude and actual noise amplitude corresponding to each key frequency point;

[0034] The predicted value correction module is used to comprehensively correct the preliminary risk prediction value according to the proportion of similar trend points and the reference noise amplitude and actual noise amplitude corresponding to each key frequency point in the set of key frequency points.

[0035] As a further limitation of the technical solution of the embodiment of the present invention, the transient test module specifically includes:

[0036] An execution record parsing unit for parsing historical execution records, screening out a predetermined number of secondary execution records that are consistent with a specific type and have no abnormalities during operation;

[0037] An amplitude range generation unit for analyzing each secondary execution record, determining its corresponding standard noise amplitude, performing denoising and smoothing processing on the obtained standard noise amplitude, and generating a standard noise amplitude range with the maximum and minimum values among the processed standard noise amplitudes as the upper and lower bounds respectively;

[0038] A test execution unit for determining whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be within the range, several transient current fluctuation tests are performed on the PCB layout of the target unit within a preset time period.

[0039] As a further limitation of the technical solution of the embodiment of the present invention, the spectrogram drawing module specifically includes:

[0040] A data acquisition unit for, during each transient current fluctuation test, using a high-speed data acquisition device to collect the test noise amplitude, the parasitic parameter change value, and the frequency range of each test in real time;

[0041] A spectrogram drawing unit for constructing a first dynamic spectrogram based on the test noise amplitude and the frequency range, and constructing a second dynamic spectrogram based on the parasitic parameter change value and the frequency range;

[0042] A spectrogram analysis unit for comparing and analyzing the first dynamic spectrogram and the second dynamic spectrogram. By calculating the correlation coefficient of the signal amplitude change between the two spectrograms in each frequency range, when the correlation coefficient exceeds a preset value, this frequency range is regarded as a similar trend point, counting the proportion of the similar trend points in the total key frequency range, and determining whether the proportion is greater than a preset threshold.

[0043] As a further limitation of the technical solution of the embodiment of the present invention, the set generation module specifically includes:

[0044] A set generation unit for parsing the first dynamic spectrogram, determining several key frequency points and their corresponding actual noise amplitudes in the first dynamic spectrogram, and forming a key frequency point set with the several key frequency points;

[0045] A reference amplitude calculation unit for parsing a predetermined number of secondary execution records, screening out several noise amplitude data that are consistent with each key frequency point, calculating the average value of the several noise amplitudes corresponding to each key frequency point, and using it as the reference noise amplitude corresponding to each key frequency point.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention collects noise data of the power supply path of the target unit in real time, establishes a standard noise amplitude range in combination with historical execution records, and uses the dynamic spectrogram generated by transient current fluctuation testing to achieve accurate identification of noise interference caused by changes in parasitic parameters of PCB wiring. Thus, a correction factor is generated by comprehensively calculating the similarity trend and amplitude deviation to dynamically correct the preliminary risk prediction value.

[0048] This dynamic correction mechanism can not only timely reflect potential risks of the electrical system in the initial stage of heavy-duty operation, improve the accuracy and real-time performance of risk assessment, but also provide a quantitative basis for system warning and intelligent operation and maintenance, effectively reduce the failure rate caused by noise interference, enhance the overall stability and security of the server chassis, optimize the maintenance strategy at the same time, reduce the operation and maintenance cost, and has significant engineering application and economic benefits. Description of the Drawings

[0049] Figure 1 It is a flowchart of the method provided by the embodiment of the present invention;

[0050] Figure 2 It is a flowchart of transient current fluctuation testing on the PCB wiring of the target unit in the method provided by the embodiment of the present invention;

[0051] Figure 3 It is a flowchart of comparing the first dynamic spectrogram with the second dynamic spectrogram in the method provided by the embodiment of the present invention;

[0052] Figure 4 It is a flowchart of parsing the first dynamic spectrogram in the method provided by the embodiment of the present invention;

[0053] Figure 5 It is a flowchart of comprehensively correcting the preliminary risk prediction value in the method provided by the embodiment of the present invention;

[0054] Figure 6 It is an application architecture diagram of the system provided by the embodiment of the present invention;

[0055] Figure 7 It is a structural block diagram of the transient test module in the system provided by the embodiment of the present invention;

[0056] Figure 8 It is a structural block diagram of the spectrogram drawing module in the system provided by the embodiment of the present invention;

[0057] Figure 9 It is a structural block diagram of the set generation module in the system provided by the embodiment of the present invention. Detailed Embodiments

[0058] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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.

[0059] Figure 1 The flowchart of the method provided by the embodiment of the present invention is shown.

[0060] Specifically, a server chassis management method based on data monitoring, the method specifically includes the following steps:

[0061] Step S100, at the starting stage when the target unit starts to execute a heavy-load operation, determine the specific type of the heavy-load operation and the real-time average noise amplitude of the power supply path of the target unit, and at the same time obtain the preliminary risk prediction value and historical execution record matching the target unit.

[0062] In the embodiment of the present invention, the target unit refers to a key module in the server chassis, and its operating state has an important impact on the stability of the whole machine. The heavy-load operation refers to the working pressure borne by the target unit when executing high-load tasks, and the "starting stage" refers to a short period after the target unit starts to execute the heavy-load operation, and this period is sufficient to reflect whether the noise amplitude of the power supply path is in a normal state.

[0063] The real-time average noise amplitude refers to the average noise amplitude obtained after processing (such as filtering and averaging operations) the noise signal data continuously collected by the sensors installed on the power supply path during this short period. This parameter is used to judge whether the noise level in the current power supply path is abnormal.

[0064] The power supply path is defined in this embodiment as the physical line that provides electrical energy for the target unit, including the power cord and related connectors, but does not include PCB traces.

[0065] The preliminary risk prediction value is a risk index calculated by using existing mature technical means (based on conventional parameters such as current, voltage, etc.), and is used as the initial evaluation basis for the potential electrical fault risk of the target unit during the heavy-load operation.

[0066] The historical execution record refers to the operation data and sensor data collected by the monitoring system and operation and maintenance logs for a long time during the heavy-load operation of the target unit, including the noise amplitude data measured on the power supply path of the target unit during the heavy-load operation, task type information, and related abnormal records.

[0067] Furthermore, the server chassis management method based on data monitoring further includes the following steps:

[0068] Step S200, parse the historical execution records, determine the standard noise amplitude range of the target unit when performing a specific type of overloaded operation, and judge whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset period.

[0069] Specifically, Figure 2 FIG. shows a flowchart of performing a transient current fluctuation test on the PCB wiring of the target unit.

[0070] Among them, parsing the historical execution records, determining the standard noise amplitude range of the target unit when performing a specific type of overloaded operation, and judging whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, performing several transient current fluctuation tests on the PCB wiring of the target unit within a preset period specifically includes the following steps:

[0071] Step S201, parse the historical execution records, and screen out a predetermined number of secondary execution records that are consistent with the specific type and have no abnormalities during operation;

[0072] Step S202, analyze each secondary execution record, determine its corresponding standard noise amplitude, perform denoising and smoothing processing on the obtained standard noise amplitude, and generate a standard noise amplitude range with the maximum and minimum values among the processed standard noise amplitudes as the upper and lower bounds respectively;

[0073] Step S203, judge whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset period.

[0074] In the embodiment of the present invention, the setting of the predetermined number is of great significance. Its purpose is to ensure that enough representative sample data is collected to reflect the stable noise characteristics of the target unit when performing a specific type of overloaded operation, and at the same time avoid introducing redundant data due to excessive record numbers, interfering with subsequent analysis. The setting of the predetermined number is generally determined based on the statistical characteristics of historical data and the actual operation situation, ensuring that the selected records can accurately reflect the noise level of the system in the normal state.

[0075] When analyzing each secondary execution record, first preprocess the original noise data, using filtering and smoothing algorithms to remove occasional noise and external interference signals. Then, use techniques such as the Fast Fourier Transform (FFT) to convert the time-domain signal into frequency-domain data, so as to extract the standard noise amplitude of the target unit under specific heavy-duty operating conditions. Next, after denoising and smoothing the standard noise amplitudes obtained in each record, take their maximum and minimum values as the upper and lower bounds respectively, thus generating a standard noise amplitude interval, which reflects the fluctuation range of the noise in the power supply path of the target unit under normal operating conditions and provides a reference standard for judging whether the real-time average noise amplitude is abnormal.

[0076] Judging whether the real-time average noise amplitude is within the above-mentioned standard noise amplitude interval is to verify whether the noise level of the current target unit's power supply path is in a normal state. Although the real-time average noise amplitude being within the standard interval indicates that there is no obvious noise interference risk on the surface of the power supply path, due to the close association between the PCB layout and the power supply path, its parasitic parameters may cause abnormal noise during heavy-duty operations due to large current fluctuations, thus bringing potential electrical risks.

[0077] Therefore, when the real-time average noise amplitude is within the standard interval, it is also necessary to perform several transient current fluctuation tests on the PCB layout of the target unit within a preset time period. The transient current fluctuation test generates short-time current fluctuations through a preset control method (for example, by combining a programmable power supply module with a preset timing controller and precisely regulating the current output within a preset time period through pulse width modulation to achieve short-time current fluctuations), and uses a high-speed data acquisition device to record the noise amplitude and the change values of parasitic parameters during the test, so as to evaluate the adverse effects that the PCB layout may have on the power supply path noise due to changes in parasitic parameters during heavy-duty operations. This step provides key data support for the subsequent drawing of the dynamic spectrogram and risk correction.

[0078] Furthermore, the server chassis management method based on data monitoring further includes the following steps:

[0079] Step S300, obtain the test noise amplitude and the change values of parasitic parameters during each transient current fluctuation test, and accordingly draw a first dynamic spectrogram and a second dynamic spectrogram, and compare whether the proportion of similar trend points in the first dynamic spectrogram and the second dynamic spectrogram is greater than a preset threshold.

[0080] Specifically, Figure 3 shows the flow chart for comparing the first dynamic spectrogram and the second dynamic spectrogram.

[0081] Among them, obtaining the test noise amplitude and the parasitic parameter variation value during each transient current fluctuation test, and accordingly plotting a first dynamic frequency spectrum diagram and a second dynamic frequency spectrum diagram, and comparing whether the proportion of similar trend points in the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram is greater than a preset threshold specifically includes the following steps:

[0082] Step S301, during each transient current fluctuation test, use a high-speed data acquisition device to collect the test noise amplitude, the parasitic parameter variation value, and the frequency range of each test in real time;

[0083] Step S302, construct a first dynamic frequency spectrum diagram according to the test noise amplitude and the frequency range, and construct a second dynamic frequency spectrum diagram according to the parasitic parameter variation value and the frequency range;

[0084] Step S303, compare and analyze the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram. By calculating the correlation coefficient of the signal amplitude change between the two spectrum diagrams in each frequency range, when the correlation coefficient exceeds a preset value, this frequency range is regarded as a similar trend point, and the proportion of the similar trend points in the total key frequency range is counted, and it is judged whether the proportion is greater than a preset threshold.

[0085] In the embodiment of the present invention, during each transient current fluctuation test, a high-speed data acquisition device is used to collect the test noise amplitude, the parasitic parameter variation value, and the frequency range of each test in real time. For this purpose, a high-precision sensor and high-speed ADC technology are used to capture the target signal, and then through preprocessing (such as filtering, denoising, and normalization) to ensure the accuracy and stability of the collected data. Next, according to the collected test noise amplitude and the frequency range, a first dynamic frequency spectrum diagram is constructed by means of digital signal processing (such as fast Fourier transform and window function processing); at the same time, using the same or similar frequency domain analysis technology, a second dynamic frequency spectrum diagram is generated according to the parasitic parameter variation value and the frequency range. These two spectrum diagrams respectively reflect the dynamic change characteristics of the test noise and the parasitic parameter variation in different frequency ranges, thus providing a reliable data basis for subsequent analysis.

[0086] Next, compare and analyze the first dynamic spectrum diagram and the second dynamic spectrum diagram. Use the Pearson correlation coefficient as a quantization index to calculate the correlation coefficient of the signal amplitude changes between the two spectrum diagrams in each frequency interval. This correlation coefficient is used to evaluate the linear correlation degree between the test noise amplitude and the parasitic parameter variation in this frequency interval. When the calculated correlation coefficient exceeds the preset value, it is considered that there is a similar trend between the two in this frequency interval, and this frequency interval is regarded as a similar trend point. Statistically analyzing the proportion of these similar trend points in the total critical frequency interval can reflect the degree of correlation of the noise fluctuations caused by the parasitic parameter changes in the PCB wiring of the target unit under heavy load operating conditions. A higher proportion of similar trend points means that in the critical frequency interval, the parasitic parameter changes have a significant impact on the noise amplitude, indicating that there may be potential electrical risks in the subsequent work.

[0087] The setting of the preset threshold is usually based on the statistical analysis of historical data and experimental calibration. By comparing the similar trend ratio under normal operating conditions with the change law under abnormal conditions, a critical value for distinguishing normal and abnormal is determined, thus providing a basis for risk correction. This ratio and its threshold not only help to identify potential hidden dangers revealed by the transient current fluctuation test of the PCB wiring when the noise level on the surface of the power supply path does not show abnormalities, but also provide quantitative data support for the subsequent risk correction process.

[0088] Furthermore, the server chassis management method based on data monitoring further includes the following steps:

[0089] Step S400, if it is determined that the proportion of similar trend points is greater than the preset threshold, then determine the set of critical frequency points based on the first dynamic spectrum diagram, and determine the reference noise amplitude and the actual noise amplitude corresponding to each critical frequency point.

[0090] Specifically, Figure 4 A flowchart for analyzing the first dynamic spectrum diagram is shown.

[0091] Among them, if it is determined that the proportion of similar trend points is greater than the preset threshold, then determining the set of critical frequency points based on the first dynamic spectrum diagram and determining the reference noise amplitude and the actual noise amplitude corresponding to each critical frequency point specifically includes the following steps:

[0092] Step S401, analyze the first dynamic spectrum diagram, determine a number of critical frequency points and their corresponding actual noise amplitudes in the first dynamic spectrum diagram, and form a set of critical frequency points with the number of critical frequency points;

[0093] Step S402, analyze a predetermined number of secondary execution records, screen out a number of noise amplitude data consistent with each critical frequency point, calculate the average value of the number of noise amplitudes corresponding to each critical frequency point, and use it as the reference noise amplitude corresponding to each critical frequency point.

[0094] In an embodiment of the present invention, when the proportion of similar trend points is greater than a preset threshold, the system first extracts key frequency points from the first dynamic spectrogram. These key frequency points are determined by analyzing the obvious peaks, stability, and noise characteristics in the spectrogram, and they represent the inherent noise characteristics during the operation of the target unit. The actual noise amplitude corresponding to each key frequency point in the first dynamic spectrogram is the noise amplitude measured in real time at that frequency point, which reflects the actual noise performance of the system under the current test conditions.

[0095] Next, by parsing a predetermined number of secondary execution records, noise amplitude data consistent with each key frequency point is screened out from the historical data. Specifically, when implementing, the noise amplitude corresponding to the key frequency point in each secondary execution record is extracted, and after denoising and smoothing these data, the average value is calculated, and this average value is defined as the reference noise amplitude corresponding to the key frequency point. The reference noise amplitude represents the standard noise level at this key frequency point when the target unit is performing a specific type of heavy-load operation and the system is in a normal operating state, providing a quantitative reference for subsequent risk assessment and risk correction.

[0096] Furthermore, the server chassis management method based on data monitoring further includes the following steps:

[0097] Step S500, according to the proportion of similar trend points and the reference noise amplitude and actual noise amplitude corresponding to each key frequency point in the set of key frequency points, comprehensively correct the preliminary risk prediction value.

[0098] Specifically, Figure 5 The flowchart showing the comprehensive correction of the preliminary risk prediction value is shown.

[0099] Among them, comprehensively correcting the preliminary risk prediction value according to the proportion of similar trend points and the reference noise amplitude and actual noise amplitude corresponding to each key frequency point in the set of key frequency points specifically includes the following steps:

[0100] Step S501, calculate and generate a first correction factor according to the proportion of similar trend points, and calculate and generate a second correction factor according to the reference noise amplitude and actual noise amplitude corresponding to each key frequency point in the set of key frequency points;

[0101] Step S502, retrieve a preset correction formula, and apply the first correction factor and the second correction factor to the preliminary risk prediction value in combination to achieve the comprehensive correction of the preliminary risk prediction value.

[0102] The correction formula is: , where refers to the corrected risk prediction value, refers to the preliminary risk prediction value, refers to the first correction factor, refers to the second correction factor;

[0103] In the correction formula, , where refers to the calibration coefficient for adjusting the influence weight of the first correction factor, refers to the proportion of similar trend points;

[0104] , where refers to the calibration coefficient for adjusting the influence weight of the second correction factor, refers to the set of key frequency points, refers to belonging to the set of key frequency points a certain key frequency point in, refers to the key frequency point the corresponding actual noise amplitude, refers to the key frequency point the corresponding reference noise amplitude, refers to the key frequency point the corresponding weight.

[0105] In the embodiment of the present invention, the first correction factor mainly reflects the proportion of similar trend points in the first dynamic spectrogram and the second dynamic spectrogram. This proportion reflects the overall correlation between the change of PCB wiring parasitic parameters and the dynamic change of the noise amplitude in the power supply path during the heavy-duty operation of the target unit. If the similar trend ratio is relatively high, it indicates that in the key frequency region, the noise change is highly consistent with the parasitic parameter change, predicting that the electrical risk may be relatively large.

[0106] The second correction factor quantifies the deviation between the actual noise amplitude at the key frequency point and the pre-determined reference noise amplitude, describing from a microscopic perspective the degree of deviation of the current noise level from the noise performance under normal working conditions.

[0107] Combining the first correction factor and the second correction factor to comprehensively correct the preliminary risk prediction value can take into account both the consistency of the overall trend and the specific amplitude deviation at the key frequency, thus more accurately reflecting the potential electrical risk caused by the change of PCB wiring parasitic parameters. The synergistic effect between the two can effectively make up for the limitations that may exist in the risk prediction of a single index, and further improve the accuracy and reliability of the entire risk assessment model.

[0108] Furthermore, Figure 6 shows the application architecture diagram of the system provided by the embodiment of the present invention.

[0109] Among them, in another preferred embodiment provided by the present invention, a server chassis management system based on data monitoring includes:

[0110] A data acquisition module 100, configured to determine the specific type of the heavy-load operation and the real-time noise average amplitude of the power supply path of the target unit at the starting stage when the target unit starts to execute the heavy-load operation, and simultaneously obtain a preliminary risk prediction value and historical execution records matching the target unit.

[0111] In the embodiment of the present invention, the target unit refers to a key module in the server chassis, and its operating state has an important impact on the stability of the whole machine. The heavy-load operation refers to the working pressure borne by the target unit when performing high-load tasks, and the "starting stage" refers to a short period after the target unit starts to execute the heavy-load operation, and this period is sufficient to reflect whether the noise amplitude of the power supply path is in a normal state.

[0112] The real-time noise average amplitude refers to the average noise amplitude obtained after processing (such as filtering and averaging operations) the noise signal data continuously collected by the sensors installed on the power supply path during this short period. This parameter is used to determine whether the noise level in the current power supply path is abnormal.

[0113] The power supply path is defined in this embodiment as the physical line that provides electrical energy for the target unit, including the power cord and related connectors, but does not include PCB traces.

[0114] The preliminary risk prediction value is a risk index calculated by using existing mature technical means (based on conventional parameters such as current and voltage), and is used as the initial evaluation basis for the potential electrical fault risk of the target unit during the heavy-load operation.

[0115] The historical execution records refer to the operation data and sensor data collected by the monitoring system and operation and maintenance logs for a long time during the heavy-load operation of the target unit, including the noise amplitude data measured on the power supply path of the target unit during the heavy-load operation, task type information, and related abnormal records.

[0116] Furthermore, the server chassis management system based on data monitoring further includes:

[0117] A transient test module 200, configured to analyze the historical execution records, determine the standard noise amplitude range when the target unit executes a specific type of heavy-load operation, and determine whether the real-time noise average amplitude is within the standard noise amplitude range. If it is determined to be within the range, several transient current fluctuation tests are performed on the PCB layout of the target unit within a preset period.

[0118] Specifically, Figure 7 shows the structural block diagram of the transient test module 200 in the system provided by the embodiment of the present invention.

[0119] Among them, in the preferred embodiment provided by the present invention, the transient test module 200 specifically includes:

[0120] An execution record parsing unit 201, configured to parse historical execution records, and screen out a predetermined number of secondary execution records that are consistent with a specific type and have no anomalies during operation;

[0121] An amplitude range generation unit 202, configured to analyze each secondary execution record, determine its corresponding standard noise amplitude, perform denoising and smoothing processing on the obtained standard noise amplitude, and generate a standard noise amplitude range with the maximum and minimum values among a number of processed standard noise amplitudes as the upper and lower bounds respectively;

[0122] A test execution unit 203, configured to determine whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be so, several transient current fluctuation tests are performed on the PCB routing of the target unit within a preset time period.

[0123] In the embodiment of the present invention, the setting of the predetermined number is of great significance. The purpose is to ensure that sufficient representative sample data is collected to reflect the stable noise characteristics of the target unit when performing specific types of heavy-duty operations, while avoiding introducing redundant data due to excessive record numbers and interfering with subsequent analysis. The setting of the predetermined number is generally determined based on the statistical characteristics of historical data and the actual operating conditions, ensuring that the selected records can accurately reflect the noise level of the system in the normal state.

[0124] When analyzing each secondary execution record, first preprocess the original noise data, use filtering and smoothing algorithms to remove accidental noise and external interference signals, and then use technologies such as fast Fourier transform (FFT) to convert the time-domain signal into frequency-domain data, so as to extract the standard noise amplitude of the target unit under specific heavy-duty operation conditions. Then, after denoising and smoothing the standard noise amplitude obtained from each record, take its maximum and minimum values as the upper and lower bounds respectively, thereby generating a standard noise amplitude range. This range reflects the fluctuation range of the noise of the power supply path of the target unit under normal working conditions, providing a reference standard for judging whether the real-time average noise amplitude is abnormal.

[0125] Judging whether the real-time average noise amplitude is within the above standard noise amplitude range is to verify whether the noise level of the current power supply path of the target unit is in a normal state. Although the real-time average noise amplitude being within the standard range indicates that there is no obvious noise interference risk on the surface of the power supply path, since the PCB routing is closely related to the power supply path, its parasitic parameters may cause abnormal noise due to large current fluctuations during heavy-duty operations, thus bringing potential electrical risks.

[0126] Therefore, when the average amplitude of the real-time noise is within the standard range, transient current fluctuation tests need to be carried out on the PCB wiring of the target unit several times within a preset period. The transient current fluctuation test generates short-term current fluctuations through a preset control method, and uses a high-speed data acquisition device to record the noise amplitude and the variation value of parasitic parameters during the test, so as to evaluate the adverse effects that the PCB wiring may have on the power supply path noise due to the change of parasitic parameters during heavy-load operation. This step provides key data support for the subsequent drawing of the dynamic spectrogram and risk correction.

[0127] Furthermore, the server chassis management system based on data monitoring further includes:

[0128] A spectrogram drawing module 300, configured to obtain the test noise amplitude and the variation value of parasitic parameters during each transient current fluctuation test, and draw a first dynamic spectrogram and a second dynamic spectrogram based on this, and compare whether the proportion of similar trend points in the first dynamic spectrogram and the second dynamic spectrogram is greater than a preset threshold.

[0129] Specifically, Figure 8 FIG. shows the structural block diagram of the spectrogram drawing module 300 in the system provided by the embodiment of the present invention.

[0130] Among them, in the preferred embodiment provided by the present invention, the spectrogram drawing module 300 specifically includes:

[0131] A data acquisition unit 301, configured to use a high-speed data acquisition device to collect the test noise amplitude, the variation value of parasitic parameters, and the frequency interval of each test in real time during each transient current fluctuation test;

[0132] A spectrogram drawing unit 302, configured to construct a first dynamic spectrogram based on the test noise amplitude and the frequency interval, and construct a second dynamic spectrogram based on the variation value of parasitic parameters and the frequency interval;

[0133] A spectrogram analysis unit 303, configured to compare and analyze the first dynamic spectrogram and the second dynamic spectrogram. By calculating the correlation coefficient of the signal amplitude change in each frequency interval of the two spectrograms, when the correlation coefficient exceeds a preset value, this frequency interval is regarded as a similar trend point, and the proportion of the similar trend points in the total key frequency interval is statistically calculated, and it is judged whether the proportion is greater than a preset threshold.

[0134] In the embodiments of the present invention, during each transient current fluctuation test, a high-speed data acquisition device is used to collect the test noise amplitude, the change value of parasitic parameters, and the frequency range of each test in real time. For this purpose, a high-precision sensor and high-speed ADC technology are used to capture the target signal, and then through preprocessing (such as filtering, denoising, and normalization) to ensure the accuracy and stability of the collected data. Next, according to the collected test noise amplitude and frequency range, a first dynamic spectrogram is constructed by means of digital signal processing (such as fast Fourier transform and window function processing); at the same time, using the same or similar frequency domain analysis techniques, a second dynamic spectrogram is generated according to the change value of parasitic parameters and the frequency range. These two spectrograms respectively reflect the dynamic change characteristics of test noise and parasitic parameter changes in different frequency ranges, thus providing a reliable data basis for subsequent analysis.

[0135] Next, compare and analyze the first dynamic spectrogram and the second dynamic spectrogram, use the Pearson correlation coefficient as a quantization index, calculate the correlation coefficient of the signal amplitude changes of the two spectrograms in each frequency range. This correlation coefficient is used to evaluate the linear correlation degree between the test noise amplitude and the change of parasitic parameters in this frequency range. When the calculated correlation coefficient exceeds the preset value, it is considered that there is a similar trend in this frequency range, and this frequency range is regarded as a similar trend point. Counting the proportion of these similar trend points in the total key frequency range can reflect the degree of correlation of the noise fluctuation caused by the change of parasitic parameters in the PCB wiring of the target unit under heavy load operating conditions. A higher proportion of similar trend points means that the change of parasitic parameters has a significant impact on the noise amplitude in the key frequency range, indicating that there may be potential electrical risks in subsequent work.

[0136] The setting of the preset threshold is usually based on the statistical analysis of historical data and experimental calibration. By comparing the similar trend ratio under normal operating conditions with the change law under abnormal conditions, a critical value for distinguishing normal and abnormal is determined, so as to provide a basis for risk correction. This ratio and its threshold not only help to confirm the potential hidden dangers revealed by the transient current fluctuation test of PCB wiring when the noise level on the surface of the power supply path does not show abnormalities, but also provide quantitative data support for the subsequent risk correction link.

[0137] Furthermore, the server chassis management system based on data monitoring further includes:

[0138] A set generation module 400, which is used to determine a set of key frequency points based on the first dynamic spectrogram and determine the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point if it is determined that the proportion of similar trend points is greater than the preset threshold.

[0139] Specifically, Figure 9The structural block diagram of the set generation module 400 in the system provided by the embodiment of the present invention is shown.

[0140] Among them, in the preferred embodiment provided by the present invention, the set generation module 400 specifically includes:

[0141] A set generation unit 401, configured to parse a first dynamic spectrogram, determine a plurality of key frequency points and their corresponding actual noise amplitudes in the first dynamic spectrogram, and form a key frequency point set with the plurality of key frequency points;

[0142] A reference amplitude calculation unit 402, configured to parse a predetermined number of secondary execution records, screen out a plurality of noise amplitude data consistent with each key frequency point, calculate the average value of the plurality of noise amplitudes corresponding to each key frequency point, and use it as the reference noise amplitude corresponding to each key frequency point.

[0143] In the embodiment of the present invention, when the proportion of similar trend points is greater than a preset threshold, the system first extracts key frequency points from the first dynamic spectrogram. These key frequency points are determined by analyzing the obvious peaks, stability, and noise characteristics in the spectrogram, and they represent the inherent noise characteristics during the operation of the target unit. The actual noise amplitude corresponding to each key frequency point in the first dynamic spectrogram is the noise amplitude measured in real time at that frequency point, which reflects the actual noise performance of the system under the current test conditions.

[0144] Next, by parsing a predetermined number of secondary execution records, noise amplitude data consistent with each key frequency point are screened out from the historical data. Specifically, in implementation, the noise amplitudes corresponding to the key frequency points in each secondary execution record are extracted, and after denoising and smoothing processing of these data, the average value is calculated, and this average value is defined as the reference noise amplitude corresponding to the key frequency point. The reference noise amplitude represents the standard noise level at this key frequency point when the target unit performs a specific type of heavy-load operation and the system is in a normal operating state, providing a quantitative reference for subsequent risk assessment and risk correction.

[0145] Furthermore, the server chassis management system based on data monitoring further includes:

[0146] A predicted value correction module 500, configured to comprehensively correct the preliminary risk prediction value according to the proportion of similar trend points and the reference noise amplitudes and actual noise amplitudes corresponding to each key frequency point in the key frequency point set.

[0147] In the embodiments of the present invention, the first correction factor mainly reflects the proportion of similar trend points in the first dynamic spectrogram and the second dynamic spectrogram. This proportion reflects the overall correlation between the changes in the parasitic parameters of the PCB wiring and the dynamic changes in the noise amplitude of the power supply path during the heavy-load operation of the target unit. If the proportion of similar trends is relatively high, it indicates that within the key frequency region, the noise changes are highly consistent with the changes in parasitic parameters, suggesting that the electrical risk may be relatively high.

[0148] The second correction factor quantifies the deviation between the actual noise amplitude at the key frequency points and the pre-determined reference noise amplitude, describing from a microscopic perspective the degree of deviation of the current noise level from the noise performance under normal operating conditions.

[0149] Combining the first correction factor and the second correction factor to comprehensively correct the preliminary risk prediction value can take into account both the consistency of the overall trend and the specific amplitude deviation at the key frequencies, thereby more accurately reflecting the potential electrical risks caused by the changes in the parasitic parameters of the PCB wiring. The synergistic effect between the two can effectively make up for the limitations that may exist in the risk prediction of a single indicator, and further improve the accuracy and reliability of the entire risk assessment model.

[0150] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0151] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0152] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0153] The above embodiments only represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

[0154] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A server chassis management method based on data monitoring, characterized in that, The method includes: At the starting stage when the target unit begins to execute the heavy-load operation, determine the specific type of the heavy-load operation and the real-time average noise amplitude of the power supply path of the target unit, and at the same time obtain the preliminary risk prediction value and historical execution record matching the target unit; Analyze the historical execution record, determine the standard noise amplitude range when the target unit executes the specific type of heavy-load operation, and judge whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset time period; Obtain the test noise amplitude and parasitic parameter variation value during each transient current fluctuation test, and accordingly draw the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram, and compare whether the proportion of similar trend points in the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram is greater than a preset threshold; If it is determined that the proportion of similar trend points is greater than the preset threshold, determine the set of key frequency points based on the first dynamic frequency spectrum diagram, and determine the reference noise amplitude and actual noise amplitude corresponding to each key frequency point; Based on the proportion of similar trend points and the reference noise amplitude and actual noise amplitude corresponding to each key frequency point in the set of key frequency points, comprehensively correct the preliminary risk prediction value.

2. The server chassis management method based on data monitoring according to claim 1, wherein, Analyze the historical execution record, determine the standard noise amplitude range when the target unit executes the specific type of heavy-load operation, and judge whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, the steps of performing several transient current fluctuation tests on the PCB wiring of the target unit within a preset time period include: Analyze the historical execution record, and screen out a predetermined number of secondary execution records that are consistent with the specific type and have no abnormalities during operation; Analyze each secondary execution record to determine its corresponding standard noise amplitude, perform denoising and smoothing processing on the obtained standard noise amplitude, and generate a standard noise amplitude range with the maximum and minimum values among the processed standard noise amplitudes as the upper and lower bounds respectively; Judge whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be yes, perform several transient current fluctuation tests on the PCB wiring of the target unit within a preset time period.

3. The server chassis management method based on data monitoring according to claim 1, characterized in that The steps of obtaining the test noise amplitude and parasitic parameter variation value during each transient current fluctuation test, and accordingly drawing the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram, and comparing whether the proportion of similar trend points in the first dynamic frequency spectrum diagram and the second dynamic frequency spectrum diagram is greater than a preset threshold include: During each transient current fluctuation test, use a high-speed data acquisition device to collect the test noise amplitude, parasitic parameter variation value, and the frequency range of each test in real time; Construct the first dynamic frequency spectrum diagram according to the test noise amplitude and frequency range, and construct the second dynamic frequency spectrum diagram according to the parasitic parameter variation value and frequency range; Compare and analyze the first dynamic spectrogram and the second dynamic spectrogram. By calculating the correlation coefficient of the signal amplitude changes between the two spectrograms in each frequency interval, when the correlation coefficient exceeds the preset value, this frequency interval is regarded as a similar trend point. Count the proportion of the similar trend points in the total key frequency interval and determine whether the proportion is greater than the preset threshold.

4. The server chassis management method based on data monitoring according to claim 2, characterized in that, If it is determined that the proportion of the similar trend points is greater than the preset threshold, the steps of determining the set of key frequency points based on the first dynamic spectrogram and determining the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point include: Analyze the first dynamic spectrogram to determine a number of key frequency points and their corresponding actual noise amplitudes in the first dynamic spectrogram, and form a set of key frequency points with the number of key frequency points. Analyze a predetermined number of secondary execution records, screen out a number of noise amplitude data consistent with each key frequency point, calculate the average value of the number of noise amplitudes corresponding to each key frequency point, and use it as the reference noise amplitude corresponding to each key frequency point.

5. The server chassis management method based on data monitoring according to claim 4, characterized in that, The steps of comprehensively correcting the preliminary risk prediction value according to the proportion of the similar trend points and the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points include: Calculate and generate a first correction factor according to the proportion of the similar trend points, and calculate and generate a second correction factor according to the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points. Retrieve the preset correction formula, and combine the first correction factor and the second correction factor and apply them to the preliminary risk prediction value to achieve the comprehensive correction of the preliminary risk prediction value.

6. The server chassis management method based on data monitoring according to claim 5, characterized in that, The correction formula is as follows: , where refers to the corrected risk prediction value, refers to the preliminary risk prediction value, refers to the first correction factor, refers to the second correction factor; In the correction formula, , where refers to the calibration coefficient for adjusting the influence weight of the first correction factor, refers to the proportion of similar trend points; , where refers to the calibration coefficient for adjusting the influence weight of the second correction factor, refers to the set of key frequency points, refers to belonging to the set of key frequency points a certain key frequency point in refers to the key frequency point corresponding actual noise amplitude, refers to the key frequency point corresponding reference noise amplitude, refers to the key frequency point corresponding weight.

7. A server chassis management system based on data monitoring, characterized in that, The system includes: a data acquisition module, a transient test module, a spectrogram drawing module, a set generation module, and a prediction value correction module, where: The data acquisition module is used to determine the real-time noise average amplitude of the specific type of the heavy-load operation and the power supply path of the target unit at the starting stage when the target unit starts to execute the heavy-load operation, and at the same time obtain the preliminary risk prediction value and the historical execution record matching the target unit. The transient test module is used to analyze the historical execution record, determine the standard noise amplitude interval when the target unit executes the specific type of heavy-load operation, and judge whether the real-time noise average amplitude is within the standard noise amplitude interval. If it is determined to be yes, perform a number of transient current fluctuation tests on the PCB wiring of the target unit within a preset time period. The spectrogram drawing module is used to obtain the test noise amplitude and the parasitic parameter change value during each transient current fluctuation test, and draw the first dynamic spectrogram and the second dynamic spectrogram based on this, and compare whether the proportion of the similar trend points in the first dynamic spectrogram and the second dynamic spectrogram is greater than the preset threshold. The set generation module is used to, if it is determined that the proportion of the similar trend points is greater than the preset threshold, determine the set of key frequency points based on the first dynamic spectrogram, and determine the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point. The prediction value correction module is used to comprehensively correct the preliminary risk prediction value according to the proportion of the similar trend points and the reference noise amplitude and the actual noise amplitude corresponding to each key frequency point in the set of key frequency points.

8. The server chassis management system based on data monitoring according to claim 7, characterized in that, The transient test module specifically includes: An execution record parsing unit, configured to parse historical execution records, and screen out a predetermined number of secondary execution records that are consistent with a specific type and have no abnormalities during operation; An amplitude range generation unit, configured to analyze each secondary execution record, determine its corresponding standard noise amplitude, perform denoising and smoothing processing on the obtained standard noise amplitude, and generate a standard noise amplitude range by using the maximum and minimum values among the processed standard noise amplitudes as the upper and lower bounds respectively; A test execution unit, configured to determine whether the real-time average noise amplitude is within the standard noise amplitude range. If it is determined to be within the range, several transient current fluctuation tests are performed on the PCB layout of the target unit within a preset time period.

9. The server chassis management system based on data monitoring according to claim 8, wherein, The spectrogram drawing module specifically includes: A data acquisition unit, configured to, during each transient current fluctuation test, use a high-speed data acquisition device to collect the test noise amplitude, the parasitic parameter change value, and the frequency range of each test in real time; A spectrogram drawing unit, configured to construct a first dynamic spectrogram based on the test noise amplitude and the frequency range, and construct a second dynamic spectrogram based on the parasitic parameter change value and the frequency range; A spectrogram analysis unit, configured to compare and analyze the first dynamic spectrogram and the second dynamic spectrogram. By calculating the correlation coefficient of the signal amplitude change between the two spectrograms in each frequency range, when the correlation coefficient exceeds a preset value, the frequency range is regarded as a similar trend point, and the proportion of the similar trend points in the total key frequency range is statistically calculated, and it is determined whether the proportion is greater than a preset threshold.

10. The server chassis management system based on data monitoring according to claim 9, characterized in that, The set generation module specifically includes: A set generation unit, configured to parse the first dynamic spectrogram, determine several key frequency points and their corresponding actual noise amplitudes in the first dynamic spectrogram, and form a key frequency point set with the several key frequency points; A reference amplitude calculation unit, configured to parse a predetermined number of secondary execution records, screen out several noise amplitude data that are consistent with each key frequency point, calculate the average value of the several noise amplitudes corresponding to each key frequency point, and use it as the reference noise amplitude corresponding to each key frequency point.

Images