Signal quality prediction method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a signal quality prediction method and device, a storage medium and electronic equipment, and the method comprises the steps: carrying out the signal quality test of a signal link based on a preset time interval, so as to determine a first signal parameter of the signal link, the first signal parameter at least comprises one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error code corresponding to the signal link and a link state change frequency corresponding to the signal link; fitting a plurality of straight lines according to a first test time for signal quality test of the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, the abscissa of a test point in each straight line being the first test time, and the ordinate being the first signal parameter; and predicting the signal quality of the signal link according to the plurality of first slopes.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and in particular, to a method and device for predicting signal quality, a storage medium, and an electronic device. Background Art

[0002] Distributed storage is a data storage technology that uses the disk space on each machine in the enterprise through the network, and these scattered storage resources form a virtual storage device, so that data is stored in every corner of the enterprise. The distributed network storage system adopts an extensible system structure, uses multiple storage servers to share the storage load, and uses location servers to locate storage information. It not only improves the reliability, availability and access efficiency of the system, but also is easy to expand. The application scenarios of distributed storage are very wide, and the requirements for the hardware reliability of storage nodes are getting higher and higher. Distributed storage is based on general servers with distributed software for management. Based on the advantages of distributed software, some complex monitoring and fault prediction of hardware reliability can be performed. At present, distributed software has relatively little monitoring of hardware reliability. It is a major trend in technology to incorporate some complex algorithms for monitoring the reliability of the entire machine into distributed software.

[0003] Current distributed storage software only performs periodic error monitoring on each Serial Attached SCSI (SAS, where SCSI stands for Small Computer System Interface) link, and only displays the monitored data on the management software interface and performs alarm processing. There is no fault prediction or processing measures, and the monitored high-speed signal links are not fully covered.

[0004] Therefore, the problem of incomplete coverage of high-speed signal link monitoring scenarios caused by only periodic bit error monitoring of the SAS link and no monitoring of other high-speed signals in the related technology has not yet been effectively solved. Summary of the invention

[0005] The embodiments of the present application provide a signal quality prediction method and device, a storage medium, and an electronic device to at least solve the problem in the related art that only periodic bit error monitoring is performed on the SAS link, and other high-speed signals are not monitored, resulting in incomplete coverage of high-speed signal link monitoring scenarios.

[0006] According to an embodiment of the present application, a signal quality prediction method is provided, which specifically includes: performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link; fitting multiple straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the horizontal coordinate of a test point in each straight line is the first test time, and the vertical coordinate is the first signal parameter; predicting the signal quality of the signal link according to the multiple first slopes.

[0007] In an exemplary embodiment, before performing a signal quality test on a signal link based on a preset time interval, the method further includes: obtaining the business volume of a server cluster where the server corresponding to the signal link is located, and determining a second test time when the signal quality test on the signal link was last performed; calculating the time interval between the second test time and the current time; determining that the signal quality test on the signal link is allowed when it is determined that the business volume is less than or equal to a business volume threshold, and the time interval is greater than or equal to a time interval threshold; determining that the signal quality test on the signal link is not allowed when it is determined that the business volume is greater than the business volume threshold, and / or the time interval is less than the time interval threshold.

[0008] In an exemplary embodiment, a signal quality test is performed on a signal link based on a preset time interval to determine a first signal parameter of the signal link, including: when it is determined that the signal link is a high-speed serial signal, controlling a first test platform to perform a signal quality test on the signal link to obtain an eye height corresponding to the signal link and an eye width corresponding to the signal link; when it is determined that the signal link is a storage link signal, controlling a second test platform to perform a signal quality test on the signal link to obtain a link error corresponding to the signal link and a number of link state changes corresponding to the signal link.

[0009] In an exemplary embodiment, predicting the signal quality of the signal link based on multiple first slopes includes: determining a first size relationship between each first slope and a preset threshold; when the first size relationship indicates that the multiple first slopes are all greater than or equal to the preset threshold, determining that the signal quality of the signal link has no degradation trend; when the first size relationship indicates that a second slope among the multiple first slopes is greater than or equal to the preset threshold, and the other slopes among the multiple first slopes are less than the preset threshold, determining a first time period corresponding to the second slope and a second time period corresponding to the other slopes, wherein the first time period includes: a first test time corresponding to a data point in a straight line corresponding to the second slope, and the second time period includes: a first test time corresponding to a data point in a straight line corresponding to the other slopes, and the other slopes are slopes among the multiple first slopes except the second slope; determining a first time point in the first time period and a second time point in the second time period A second size relationship of a second time point in a time period, wherein the first time point is a time point corresponding to any data point in the first time period except the time point corresponding to the last data point in the straight line corresponding to the second slope, and the second time point is a time point corresponding to any data point in the straight lines corresponding to the other slopes; when the second size relationship indicates that the first time point is greater than the second time point, a second signal quality test is performed on the signal link to determine the second signal parameter of the signal link, and a second prediction is performed on the signal quality of the signal link based on the second signal parameter to determine the signal quality of the signal link based on the first prediction result; when the second size relationship indicates that the first time point is greater than the second time point, it is determined that the signal quality of the signal link has no deterioration trend; when the first size relationship indicates that the multiple first slopes are all less than the preset threshold, it is determined that the signal quality of the signal link has a deterioration trend.

[0010] In an exemplary embodiment, after predicting the signal quality of the signal link according to multiple first slopes, the method also includes: when it is determined that the signal quality of the signal link has a deterioration trend, determining whether the first signal parameter is greater than or equal to a signal parameter threshold; when it is determined that the first signal parameter is less than the signal parameter threshold, starting a signal link alarm mechanism and recording the number of alarms of the signal link within a third time period; when the number of alarms within the third time period is less than or equal to the alarm number threshold, determining that the signal quality of the signal link is updated from having a deterioration trend to having no deterioration trend; when the number of alarms within the third time period is greater than the alarm number threshold, updating the preset time interval so that the updated preset time interval is less than the preset time interval; performing a third signal quality test on the signal link based on the updated preset time interval to determine the third signal parameter of the signal link, and performing a third prediction on the signal quality of the signal link based on the third signal parameter to determine the signal quality of the signal link according to the second prediction result.

[0011] In an exemplary embodiment, after determining whether the first signal parameter is greater than or equal to a signal parameter threshold, the method further includes: when it is determined that the first signal parameter is greater than or equal to the signal parameter threshold, performing fault elimination processing on the signal link; when it is determined that the fault corresponding to the signal link has been eliminated, updating the signal quality of the signal link from having a degradation trend to having no degradation trend.

[0012] In an exemplary embodiment, troubleshooting the signal link includes: troubleshooting the signal link in one of the following ways: troubleshooting the signal link by restarting the physical layer interface corresponding to the signal link; troubleshooting the signal link by isolating the server corresponding to the signal link from the server cluster where the server is located, and restarting the isolated server; troubleshooting the signal link by reducing the corresponding operating rate of the server; troubleshooting the signal link by reducing the data transmission bandwidth corresponding to the signal link.

[0013] According to another embodiment of the present application, a signal quality prediction device is provided, including: a testing module, used to perform a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link; a fitting module, used to fit multiple straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determine a first slope corresponding to each straight line, wherein the horizontal coordinate of a test point in each straight line is the first test time, and the vertical coordinate is the first signal parameter; a prediction module, used to predict the signal quality of the signal link according to multiple first slopes.

[0014] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps in the above method embodiment when running.

[0015] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in the above method embodiment.

[0016] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0017] According to the signal quality prediction method of the present application, the signal quality test of the signal link is performed based on a preset time interval to determine the first signal parameter of the signal link, and the first signal parameter can be: the eye height corresponding to the signal link, the eye width corresponding to the signal link, the link error corresponding to the signal link, and the number of link state changes corresponding to the signal link; according to the first test time and the first signal parameter of the signal quality test of the signal link, multiple straight lines are fitted, wherein the abscissa of the data point in each straight line is the first test time, and the ordinate is the first signal parameter, and the first slope of each straight line is determined; the signal quality of the signal link is tested according to the multiple first slopes of the multiple straight lines. That is, the embodiment of the present application determines the first signal parameter corresponding to the signal link by testing the signal link, and the first signal parameter can include not only link error, but also eye height, eye width, number of link state changes, etc., that is, it is not limited to the signal quality test of the SAS link, and the signal quality test can also be performed on other signal links; according to the first test time and the first signal parameter of the signal quality test, multiple straight lines are fitted, and then the first slope of each straight line is determined, and the signal quality of the signal link can be predicted according to the first slope. Through the embodiments of the present application, the problem of incomplete coverage of high-speed signal link monitoring scenarios caused by only periodic bit error monitoring of SAS links in the related technology and no monitoring of other high-speed signals can be solved, thereby covering the high-speed signal link monitoring scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal of a signal quality prediction method according to an embodiment of the present application;

[0019] Figure 2 is a flow chart of a method for predicting signal quality according to an embodiment of the present application;

[0020] Figure 3 It is a system framework diagram for predicting and repairing signal quality faults of high-speed signal links suitable for distributed storage according to an optional embodiment of the present application;

[0021] Figure 4 is a schematic diagram of fitting two groups of straight lines according to an optional embodiment of the present application;

[0022] Figure 5 is a flow chart of a method for data processing and prediction according to an optional embodiment of the present application;

[0023] Figure 6 It is a structural block diagram of a signal quality prediction device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal of a signal quality prediction method according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the interactive state in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NlC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0029] Figure 2 is a flowchart of a signal quality prediction method according to an embodiment of the present application, which can be applied to Figure 1 In a computer terminal, such as Figure 2 As shown, the process includes the following steps:

[0030] Step S202, performing a signal quality test on the signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link;

[0031] Among them, the above-mentioned signal links can be: Peripheral Component Interconnect Express (PCle for short), SAS / Serial ATA (SATA for short, ATA is Advanced Technology Attachment), Unified Platform Interconnect (UPl for short), Gigabit Media Independent Interface (xGMII for short) and other signal links.

[0032] Step S204, fitting a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter;

[0033] The plurality of straight lines may be two straight lines, that is, after obtaining three test points, two straight lines may be fitted, and the first slopes respectively corresponding to the two straight lines may be obtained.

[0034] For example: based on the signal link, at time a1, the result of the first signal parameter tested is c1, at time a2, the result of the first signal parameter tested is c2, and at time a3, the result of the first signal parameter tested is c3. The test points obtained are A(a1, c1), B(a2, c2), and C(a3, c3). According to the A and B test points, the first straight line y=k1x+b1 can be obtained, and according to the B and C points, the second straight line y=k2x+b2 can be obtained. Then the above first slope is: k1 and k2.

[0035] Step S206: predicting the signal quality of the signal link according to the multiple first slopes.

[0036] According to the signal quality prediction method of the present application, the signal quality test of the signal link is performed based on a preset time interval to determine the first signal parameter of the signal link, and the first signal parameter can be: the eye height corresponding to the signal link, the eye width corresponding to the signal link, the link error corresponding to the signal link, and the number of link state changes corresponding to the signal link; according to the first test time and the first signal parameter of the signal quality test of the signal link, multiple straight lines are fitted, wherein the abscissa of the data point in each straight line is the first test time, and the ordinate is the first signal parameter, and the first slope of each straight line is determined; the signal quality of the signal link is tested according to the multiple first slopes of the multiple straight lines. That is, the embodiment of the present application determines the first signal parameter corresponding to the signal link by testing the signal link, and the first signal parameter can include not only link error, but also eye height, eye width, number of link state changes, etc., that is, it is not limited to the signal quality test of the SAS link, and the signal quality test can also be performed on other signal links; according to the first test time and the first signal parameter of the signal quality test, multiple straight lines are fitted, and then the first slope of each straight line is determined, and the signal quality of the signal link can be predicted according to the first slope. Through the embodiments of the present application, the problem of incomplete coverage of high-speed signal link monitoring scenarios caused by only periodic bit error monitoring of SAS links in the related technology and no monitoring of other high-speed signals can be solved, thereby covering the high-speed signal link monitoring scenarios.

[0037] Optionally, before performing a signal quality test on the signal link based on a preset time interval in the above-mentioned step S202, the method further includes: obtaining the business volume of the server cluster where the server corresponding to the signal link is located, and determining a second test time when the signal quality test on the signal link was last performed; calculating the time interval between the second test time and the current time; when it is determined that the business volume is less than or equal to a business volume threshold, and the time interval is greater than or equal to a time interval threshold, determining that a signal quality test on the signal link is allowed; when it is determined that the business volume is greater than the business volume threshold, and / or the time interval is less than the time interval threshold, determining that a signal quality test on the signal link is not allowed.

[0038] It is understandable that before performing a signal quality test on a signal link, it is necessary to determine the monitoring scenario of the signal link, judge the monitoring scenario, and determine whether it meets the monitoring conditions, so that the signal quality test can be performed efficiently and timely without affecting the normal business operation of the server cluster. Specifically:

[0039] (1) Obtaining business volume: Before conducting a signal quality test, you first need to collect the current business volume information of the server or server cluster corresponding to the signal link, which may include indicators such as server load, data transmission rate, and transaction volume. The acquisition of business volume is directly related to whether the signal quality test will have a negative impact on system performance.

[0040] (2) Determine the second test time: record the second test time of the last signal quality test on this signal link, and then calculate the time interval from the last test to the current time, so as to determine whether it is the right time for the next test.

[0041] (3) Calculating the time interval: By comparing the current time with the second test time, the interval between the current time and the last test time is calculated. The time interval calculation can be used to determine whether the time-based test condition is met, that is, whether the preset time interval threshold is exceeded.

[0042] For example, the time interval threshold between the current test time and the second test time can be set to 12 hours. When the time interval is determined to be 14 hours, it can be determined that the time interval is greater than the time interval threshold, and thus the time-based test condition is determined to be met.

[0043] (4) Determine the traffic volume and time interval: Set the traffic volume threshold and time interval threshold as the trigger conditions for the test. If the current traffic volume is less than or equal to the traffic volume threshold, it indicates that the load of the server cluster is relatively low, and the signal quality test will not have a significant impact on the service. At the same time, if the time interval from the last test to the current one is greater than or equal to the time interval threshold, it indicates that enough time has passed since the last test, and the signal link status may have changed, and a new round of testing is needed to ensure signal quality.

[0044] (5) Determine test authority: If both the traffic volume and the time interval meet the conditions, the signal quality test of the signal link will be allowed. On the contrary, if either or both of the traffic volume and the time interval do not meet the conditions, it will be determined that performing the signal quality test at this time may cause unnecessary interference to the operation of the server or server cluster, and thus the test will not be allowed.

[0045] The above technical solution ensures that the signal quality test will neither cause performance degradation during business peaks nor occupy system resources too frequently. Instead, it is performed under conditions where the business volume is low and the interval between the last test is long enough to achieve a balance between maintaining the health of the signal link and not affecting the normal operation of the system. Through the above pre-test judgment process, the management software of the distributed storage system or server cluster can intelligently arrange signal quality tests to avoid repeated tests under high business pressure and in a short period of time, thereby improving resource utilization efficiency and the overall stability of the system. This helps to minimize interference with the business while maintaining the health of the signal link.

[0046] Optionally, the above-mentioned step S202 performs a signal quality test on the signal link based on a preset time interval to determine a first signal parameter of the signal link, including: when it is determined that the signal link is a high-speed serial signal, controlling the first test platform to perform a signal quality test on the signal link to obtain an eye height corresponding to the signal link and an eye width corresponding to the signal link; when it is determined that the signal link is a storage link signal, controlling the second test platform to perform a signal quality test on the signal link to obtain a link error corresponding to the signal link and a number of link state changes corresponding to the signal link.

[0047] It is understandable that, when it is determined that the current environment meets the signal quality test environment, the signal quality test can be performed on the signal link. The first step in performing the signal quality test is to determine the first signal parameter of the signal link, specifically:

[0048] (1) Signal quality test for high-speed serial signal links:

[0049] The high-speed serial signal link may be a signal link such as PCIe and xGMII. The eye height and eye width of the eye diagram may be captured by calling a related test tool.

[0050] An eye diagram is a graphical representation of the waveform of a high-speed signal at the receiving end.

[0051] Among them, the eye height reflects the amplitude of the signal in the signal link, and the eye width shows the timing margin of the signal in the signal link. The eye height and eye width are directly related to the integrity and anti-interference ability of the signal in the signal link.

[0052] (2) For the signal quality test of the storage link signal:

[0053] The storage link signal may be a SAS / SATA signal link.

[0054] The link bit error rate (BER) and the number of link status changes corresponding to the signal link can be obtained by calling a test tool such as scrutiny (a test tool).

[0055] Among them, the link bit error rate is a key indicator for evaluating the error rate during data transmission, and the number of link status changes reflects the stability of the link. Frequent status changes may indicate potential hardware or configuration problems.

[0056] That is, when performing signal quality tests on two different types of signal links, the appropriate test platform and method can be selected according to the characteristics of the signal link. By controlling the first test platform and the second test platform, the first signal parameters of the corresponding types can be obtained for the characteristics of the high-speed serial signal and the storage link signal, respectively, providing an accurate data basis for subsequent signal quality analysis and fault prediction. The above technical solution ensures that in a distributed storage system, whether it is a high-speed data transmission link or a storage device link, its signal quality can be effectively monitored and evaluated, so as to identify problems in advance, take preventive or repair measures, and ensure the overall stability and reliability of the system.

[0057] Optionally, the above-mentioned step S206 predicts the signal quality of the signal link according to multiple first slopes, including: determining a first size relationship between each first slope and a preset threshold; when the first size relationship indicates that the multiple first slopes are all greater than or equal to the preset threshold, determining that the signal quality of the signal link has no degradation trend; when the first size relationship indicates that the second slope among the multiple first slopes is greater than or equal to the preset threshold, and the other slopes among the multiple first slopes are less than the preset threshold, determining a first time period corresponding to the second slope and a second time period corresponding to the other slopes, wherein the first time period includes: a first test time corresponding to a data point in a straight line corresponding to the second slope, and the second time period includes: a first test time corresponding to a data point in a straight line corresponding to the other slopes, and the other slopes are slopes of the multiple first slopes except the second slope; determining a first time point in the first time period and a second time period corresponding to the other slopes a second size relationship between second time points in two time periods, wherein the first time point is a time point corresponding to any data point in the first time period except the time point corresponding to the last data point in the straight line corresponding to the second slope, and the second time point is a time point corresponding to any data point in the straight lines corresponding to the other slopes; when the second size relationship indicates that the first time point is greater than the second time point, performing a second signal quality test on the signal link to determine a second signal parameter of the signal link, and performing a second prediction on the signal quality of the signal link based on the second signal parameter to determine the signal quality of the signal link based on the first prediction result; when the second size relationship indicates that the first time point is greater than the second time point, determining that the signal quality of the signal link has no deterioration trend; when the first size relationship indicates that the multiple first slopes are all less than the preset threshold, determining that the signal quality of the signal link has a deterioration trend.

[0058] It is understandable that the signal quality of the signal link can be divided into two types: one with a degradation trend and one without a degradation trend. Each of the above first slopes reflects the variation trend of the first signal parameter of the signal link over the first test time. The above preset threshold can be 0.

[0059] The signal quality of the signal link can be predicted based on the first slope of each straight line, specifically:

[0060] Assume that based on the signal link, at time a1, the result of the first signal parameter tested is c1, at time a2, the result of the first signal parameter tested is c2, and at time a3, the result of the first signal parameter tested is c3. The test points obtained are A(a1, c1), B(a2, c2), and C(a3, c3). According to the A and B test points, the first straight line y=k1x+b1 can be obtained, and according to the B and C points, the second straight line y=k2x+b2 can be obtained. Then the above first slope is: k1 and k2.

[0061] (1) When all first slopes are greater than or equal to a preset threshold (i.e., k1 is greater than or equal to a preset threshold.

[0062] and k2 is greater than or equal to the preset threshold):

[0063] That is, the first signal parameter does not drop significantly during the monitored time period, and the signal quality of the signal link can be determined to have no degradation trend.

[0064] (2) When a second slope among the plurality of first slopes is greater than or equal to a preset threshold, and the other slopes are less than the preset threshold (i.e., k1 is greater than or equal to the preset threshold, and k2 is less than the preset threshold; or k1 is less than the preset threshold, and k2 is greater than or equal to the preset threshold):

[0065] It is necessary to compare the second size relationship between the first time period corresponding to the test point in the straight line corresponding to the second slope and the second time period corresponding to the test point in the straight line corresponding to the other slopes. When the first time period is greater than the second time period, it means that k1 (the second slope) is greater than or equal to the preset threshold value, and k2 (the other slopes) is less than the preset threshold value.

[0066] At this time, it is necessary to perform a second signal quality test on the signal link to obtain updated signal parameters (i.e., second signal parameters). Based on the second signal parameters, the above steps S204 and S206 are repeatedly performed to perform a second signal quality prediction on the signal link to further confirm the signal quality trend of the signal link.

[0067] When the first time period is shorter than the second time period, it means that k1 (other slope) is greater than or equal to the preset threshold, and k2 (second slope) is less than the preset threshold.

[0068] At this time, the signal quality of the signal link can be determined to have no degradation trend.

[0069] (3) When all first slopes are less than the preset threshold (i.e., k1 is less than the preset threshold and k2 is less than the preset threshold):

[0070] That is, the first signal parameter decreases significantly during the monitored time period, and the signal quality of the signal link can be determined to have a deterioration trend.

[0071] Through the above technical solution, multiple first slopes can be used to dynamically monitor and predict the signal quality trend based on the changes in time series signal parameters. In addition, the degradation trend of the signal link signal quality can be identified in a timely manner, and the link status can be re-evaluated when the signal quality begins to recover, avoiding unnecessary over-monitoring or misjudgment. The preset threshold is set based on the signal parameter range and acceptable performance degradation level when the link is operating normally. By comparing with the first slope, the changing trend of the signal quality can be quantified, and scientific predictions and decisions can be made.

[0072] The dynamic signal quality prediction mechanism of the embodiment of the present application is particularly important for large systems such as distributed storage systems and high-performance computing clusters. It helps to warn of possible failures of the signal link in advance and take appropriate preventive measures. For example: adjust the signal rate, add redundant paths, or start a fault recovery process to ensure the stability, reliability and efficiency of system data transmission. By intelligently analyzing the changing trends of signal parameters, the health status of the signal link can be actively managed to reduce potential system interruptions and data loss risks.

[0073] Among them, after predicting the signal quality of the signal link according to multiple first slopes, the method also includes: when it is determined that the signal quality of the signal link has a deterioration trend, determining whether the first signal parameter is greater than or equal to the signal parameter threshold; when it is determined that the first signal parameter is less than the signal parameter threshold, starting the signal link alarm mechanism and recording the number of alarms of the signal link in a third time period; when the number of alarms in the third time period is less than or equal to the alarm number threshold, determining that the signal quality of the signal link is updated from having a deterioration trend to having no degradation trend; when the number of alarms in the third time period is greater than the alarm number threshold, updating the preset time interval so that the updated preset time interval is less than the preset time interval; performing a third signal quality test on the signal link based on the updated preset time interval to determine the third signal parameter of the signal link, and performing a third prediction on the signal quality of the signal link based on the third signal parameter to determine the signal quality of the signal link according to the second prediction result.

[0074] It is understandable that after detecting that the signal quality of the signal link has a deteriorating trend, it is necessary to further evaluate the signal parameter status and adjust the monitoring strategy to more accurately predict and manage the signal quality of the signal link. Specifically:

[0075] (1) Preliminary identification of signal quality degradation trend: Signal quality prediction is performed based on multiple first slopes. If all slopes are less than a preset threshold, it indicates that the signal parameters show a downward trend over time and the signal quality of the signal link tends to deteriorate.

[0076] (2) Signal parameter threshold comparison: After initially determining that the signal link has a degradation trend, it is possible to further check whether the first signal parameter (e.g., eye height, eye width, link bit error rate, etc.) exceeds the signal parameter threshold. If the first signal parameter is lower than the signal parameter threshold, it means that the signal quality is still within an acceptable range, despite the degradation trend.

[0077] (3) Start the alarm mechanism and record the number of alarms: If the first signal parameter is lower than the signal parameter threshold, that is, the signal quality begins to be lower than the normal level, the signal link alarm mechanism will be started to record the abnormal state of the signal link. In the third time period (for example: a few hours to a few days), the number of alarms can be continuously monitored, that is, the number of times the signal quality of the signal link is lower than the threshold during the monitoring period.

[0078] (4) Alarm threshold determination: The alarm number threshold may be 0. If the alarm number in the third time period is less than or equal to the alarm number threshold, it indicates that despite the degradation trend, the signal quality of the signal link has not degraded to the extent that emergency intervention is required. In this case, it can be determined that the signal quality of the signal link has been updated from having a degradation trend to having no degradation trend, that is, it is considered that the signal link can be recovered autonomously or the degradation trend of the signal quality of the signal link is temporary.

[0079] However, if the number of alarms exceeds the alarm number threshold, it indicates that the signal quality of the signal link is seriously degrading and more intensive monitoring and possible fault repair measures are required.

[0080] (5) Adjust monitoring frequency: When the number of alarms exceeds the alarm threshold, the preset time interval can be adjusted to shorten the preset time interval so that the signal quality of the signal link can be tested more frequently. The updated preset time interval will ensure that any changes in signal quality are captured in a timely manner.

[0081] (6) Third signal quality test and prediction: Based on the shortened preset time interval (i.e., the updated preset time interval), a third signal quality test is performed to obtain a third signal parameter. Then, a third prediction is performed based on the third signal parameter to evaluate whether the signal quality of the signal link continues to deteriorate or has returned to normal. Through the above technical solution, the latest signal link health status information of the signal link is provided, which helps the system make more accurate fault prediction and management decisions.

[0082] Through the embodiments of the present application, not only can the degradation trend of the signal link signal quality be identified, but also the monitoring strategy can be adjusted according to the real-time signal parameter status to adapt to the changes in the health status of the signal link. This dynamic monitoring and prediction mechanism helps to take measures as soon as the signal quality begins to decline, avoiding the significant impact of signal link failures on system performance and reliability, while also reducing unnecessary resource consumption and achieving more efficient and intelligent signal link management.

[0083] Among them, after determining whether the first signal parameter is greater than or equal to the signal parameter threshold, the method also includes: when it is determined that the first signal parameter is greater than or equal to the signal parameter threshold, performing fault elimination processing on the signal link; when it is determined that the fault corresponding to the signal link has been eliminated, updating the signal quality of the signal link from having a degradation trend to having no degradation trend.

[0084] It is understandable that, during the third signal quality test, if it is determined that the first signal parameter is greater than or equal to the signal parameter threshold, it is necessary to perform fault elimination processing on the signal link. Specifically:

[0085] The signal link is subjected to troubleshooting, comprising: troubleshooting the signal link in one of the following ways: troubleshooting the signal link by restarting the physical layer interface corresponding to the signal link; troubleshooting the signal link by isolating the server corresponding to the signal link from the server cluster where the server is located, and restarting the isolated server; troubleshooting the signal link by reducing the corresponding operating rate of the server; troubleshooting the signal link by reducing the data transmission bandwidth corresponding to the signal link.

[0086] It is understood that the methods for troubleshooting the signal link include:

[0087] (1) Physical layer interface restart: You can try to restart the physical layer interface (PHY) corresponding to the signal link. PHY is the hardware layer of the signal link and is responsible for physical layer operations such as signal encoding, decoding, transmission, and reception. Restarting PHY is intended to clear possible hardware-level error states and reinitialize the PHY module to restore the normal operation of the signal link. For example, for PCIe or SAS / SATA signal links, restarting PHY can re-perform link training and status negotiation, which helps to solve temporary signal transmission problems.

[0088] (2) Server isolation and restart: In more serious cases, it may be necessary to isolate the problematic server from the server cluster to which it belongs. The purpose of isolation is to prevent potential signal link problems from affecting the overall performance and stability of the cluster. Once the server is isolated, the system can safely restart the server without affecting other servers in the cluster. Restarting the server can reset the status of all its hardware components, including the PHY of the signal link, thereby troubleshooting the signal link. After restarting, the system will retest the signal link to ensure that the signal quality returns to normal, and then reintegrate the server into the cluster to resume normal service.

[0089] (3) Reduce the server operating rate: For some signal link problems, especially those related to signal integrity, reducing the server operating rate (for example, reducing the PCIe rate level from Gen4 to Gen3) may help reduce the load on the signal link, thereby reducing the signal transmission error rate. By reducing the rate, the system can temporarily sacrifice some performance in exchange for improved signal link stability and signal quality, avoiding more serious failures.

[0090] (4) Reduce data transmission bandwidth: You can also reduce the data transmission bandwidth of the signal link, which usually involves reducing the number of physical channels in the link (for example: PCIe from x8 to x4). Reducing the bandwidth can reduce the complexity of the signal link, reduce signal integrity issues such as crosstalk and signal reflection, and help maintain at least partial data transmission capacity when there is a potential hardware failure or improper configuration of the link until more thorough repair measures are implemented.

[0091] The above fault elimination processing strategies are hierarchical, ranging from the lighter software-level restart of PHY, to the heavier hardware-level restart of the server, and then to adjusting the operating parameters of the server and signal link to adapt to the current state of the signal link. The selection and implementation of these strategies are based on the evaluation of the type and severity of the signal link fault, aiming to restore or improve the signal quality of the signal link with minimal performance loss and resource consumption, ensuring the stable operation and data transmission reliability of large systems such as distributed storage systems and server clusters.

[0092] The above-mentioned technical solution for fault elimination processing can not only respond to and handle signal link faults in a timely manner, but also take the most appropriate recovery measures according to the specific nature of the fault. This multi-level fault handling mechanism combines the fault recovery capabilities of software and hardware, providing a comprehensive solution for the maintenance and management of signal links, helping to reduce maintenance costs and improve system availability and efficiency.

[0093] In order to better understand the process of the above-mentioned signal quality prediction method, the implementation process of the above-mentioned signal quality prediction method is described below in combination with an optional embodiment, but it is not used to limit the technical solution of the embodiment of the present application.

[0094] The optional embodiment of the present application relates to the field of distributed storage technology, and defines a method and system for predicting and repairing signal quality failures of a high-speed signal link applicable to distributed storage. The optional embodiment of the present application aims to improve the stability and reliability of the distributed storage system by monitoring the status information of the high-speed signal link in real time, predicting potential signal quality failures, and automatically repairing them when they occur.

[0095] Figure 3 is a system framework diagram for predicting and repairing signal quality faults of high-speed signal links suitable for distributed storage according to an optional embodiment of the present application, such as Figure 3 As shown, a system for predicting and repairing signal quality faults in high-speed signal links suitable for distributed storage includes: a high-speed link monitoring scenario judgment module, a signal quality data capture module, a data processing and prediction module, and a fault repair module. Among them:

[0096] (1) High-speed link monitoring scenario judgment module, used to:

[0097] By monitoring the business volume of the cluster (i.e., the server cluster of this application), it is ensured that the business volume of the node (i.e., the server of this application) is low, and the time interval from the last high-speed signal quality detection is greater than 6 hours; this ensures that the node performs high-speed signal detection under a lower load, and the interval is greater than 6 hours, ensuring that signal quality detection is not performed too much, occupying hardware resources.

[0098] (2) Signal quality data capture module, used for:

[0099] After entering the high-speed link monitoring process, the inspection items cover signal links such as PCIe, UPI, xGMII, SAS / SATA, etc. By calling relevant test tools, the eye height and eye width of the eye diagram are captured, and they are matched one by one with the capture time, which is convenient for subsequent data processing.

[0100] 1) PCIe, UPl, xGMII high-speed signal inspection method:

[0101] Signal quality testing can be performed using relevant test tools, namely:

[0102] Confirm the LAN number (Local Area Network, i.e. the logical address number on the network adapter or network interface card) and DIE number (a code for a specific processing unit or functional block inside a chip) of the target device (i.e. the server of this application).

[0103] Table 1 is the eye diagram corresponding to the PCIe, UPI, and xGMII high-speed signal links of an optional embodiment of the present application.

[0104] That is, through the instruction: {. / hgt eye-s-dxx, xx}, each DIE number of the central processing unit (CPU) in the server is traversed, that is, DIE0~7 needs to be traversed to search for all PCIe and xGMII signals, and then the eye diagram shown in Table 1 can be collected and drawn.

[0105] Table 1

[0106]

[0107]

[0108] The eye height and eye width (ie, the first signal parameter) of the signal link can be determined by using the above test tool.

[0109] 2) Signal quality collection method of SAS / SATA signal link:

[0110] Relevant testing tools may be used to perform signal quality testing. Table 2 is a schematic diagram of signal parameters corresponding to a SAS / SATA signal link according to an optional embodiment of the present application.

[0111] That is, through the command: {scrtnycli.x86_64-i[x]phy-err}, query the link error of PHY; through the command: {scrtnycli.x86_64-i[index]cliphyinfo}, query the change Count (that is, the number of link status changes) of PHY;

[0112] Table 2

[0113]

[0114]

[0115] By using the above test tool, the link bit errors and the number of link status changes (ie, the first signal parameter) of the signal link can be determined.

[0116] (3) Data processing and prediction module, used for:

[0117] The test results of the three sets of data read cumulatively are all Pass.

[0118] Figure 4 is a schematic diagram of fitting two sets of straight lines according to an optional embodiment of the present application, such as Figure 4As shown: Two sets of straight lines are fitted according to time and data, and three sets of data with coordinates A (a0, b0), B (c0, dO) and B (c0, dO), C (e0, f0) (the above A, B, C are the test points) form two sets of linear equations of two variables: y = k1x + b1, k1 is the slope of the straight line, b1 is the vertical intercept and y = k2x + b2, k2 is the slope of the straight line, b2 is the vertical intercept.

[0119] Figure 5 is a flow chart of a method for data processing and prediction according to an optional embodiment of the present application, such as Figure 5 As shown:

[0120] Step S501, determine whether k1 and k2 are both less than 0.

[0121] If it is determined that both k1 and k2 are less than 0, step S502 is executed; otherwise, step S508 is executed.

[0122] That is, if the slopes of the two lines k1 and k2 are both less than 0, it means that the LAN signal has a degradation trend; if k1 is less than 0 and k2 is greater than 0, the link has no degradation trend, the link is normal, and all three sets of data are cleared, and the next three sets of data are accumulated; if k1 is greater than or equal to 0 and k2 is greater than or equal to 0, the link has no degradation trend and the link is normal. If k1 is greater than or equal to 0 and k2 is less than 0, the link has no degradation trend and the link is normal, the data of coordinate A is discarded, the original coordinate B data is retained and revised to A coordinate data, the original coordinate C data is retained and revised to B coordinate data, and the next set of query result data is continued to define coordinate data C.

[0123] Step S502: determine whether a signal parameter (ie, a first signal parameter) exceeds a threshold (ie, a signal parameter threshold).

[0124] If it is determined that the threshold is exceeded, step S503 is executed; if it is determined that the threshold is not exceeded, step S505 is executed.

[0125] Step S503: alarm and fault handling.

[0126] If the threshold is exceeded, an alarm link abnormality is output and the fault handling module is entered.

[0127] Step S504, determining whether the alarm is eliminated.

[0128] If it is determined that the error has been eliminated, the link is determined to be normal and step S508 is executed. If it is determined that the error has not been eliminated, the link is determined to be abnormal and step S509 is executed.

[0129] Step S505, start the alarm mechanism (the link has a tendency to deteriorate).

[0130] Step S506, accumulating the number of alarms.

[0131] Step S507, determining whether the alarm number is 0 (ie, the alarm number threshold).

[0132] When it is determined that the number of alarms is 0, it is determined that the link is normal, and step S508 is executed; when it is determined that the number of alarms is not 0, the signal quality is captured again, and the data is processed and predicted, and step S501 is executed.

[0133] That is, if it does not exceed the threshold, the alarm link has a degradation trend, and the cumulative alarm times are calculated. If the alarm times are not 0, the high-density monitoring process is entered, and monitoring is performed once every 1 hour. If there is no downward trend, the cumulative alarm value is directly cleared to 0, and the output link is normal. Based on the second set of straight lines, the signal quality in the next time period can be predicted, and it can be predicted that it may degrade to the set threshold at some time in the future, and the predicted time point of exceeding the threshold is output.

[0134] Step S508: the output link is normal.

[0135] Step S509: output link abnormality.

[0136] In case of an output link failure, (4) a fault recovery module is used to:

[0137] If the high-speed signal link exceeds the threshold, the following strategies can be used to repair it according to the business scenario:

[0138] 1) Restart the faulty PHY: When a faulty PHY (physical layer interface) is detected, first try to restart the faulty PHY to see if it can be repaired after restart.

[0139] 2) Speed ​​reduction: Determine the actual operating rate of the device and whether it meets the demand after the speed reduction. If it does, a bandwidth reduction strategy can be implemented, for example, reducing Gen4 to Gen3.

[0140] 3) Reduce bandwidth: Determine the actual operating rate of the device and whether it meets the requirements after reducing the bandwidth. If it does, a bandwidth reduction strategy can be implemented, such as reducing X8 to X4.

[0141] 4) Restart the machine: The distributed storage cluster has a mechanism that can fault a node. When the business pressure is relatively small, you can start to isolate a node, restart the isolated node, retrain the high-speed signal, debug the optimal parameters to start, and try to repair the faulty PHY; after the repair is complete, the node will continue to be expanded into the cluster and continue to be used.

[0142] The optional embodiment of the present application monitors the high-speed signals of each distributed node under the cluster, such as PCle, UPI, xGMII, SAS / SATA links, etc., and promptly warns whether the signal has a degradation trend, and can predict the time point when the threshold may be exceeded, with the ability to predict in advance. When a fault occurs, the system can autonomously perform repair operations without manual intervention, reducing maintenance costs and improving system stability, repairing potential faults in a timely manner, avoiding fault expansion, and improving the overall stability and reliability of the distributed storage system.

[0143] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0144] In this embodiment, a signal quality prediction device is also provided, which is used to implement the above embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0145] Figure 6 is a structural block diagram of a signal quality prediction device according to an embodiment of the present application, such as Figure 6 As shown, the device comprises:

[0146] A testing module 62 is configured to perform a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link;

[0147] A fitting module 64, configured to fit a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determine a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter;

[0148] The prediction module 66 is configured to predict the signal quality of the signal link according to the multiple first slopes.

[0149] Through the signal quality prediction device of the present application, the signal quality test is performed on the signal link based on the preset time interval to determine the first signal parameter of the signal link, and the first signal parameter can be: the eye height corresponding to the signal link, the eye width corresponding to the signal link, the link error corresponding to the signal link, and the number of link state changes corresponding to the signal link; according to the first test time and the first signal parameter of the signal quality test of the signal link, multiple straight lines are fitted, wherein the abscissa of the data point in each straight line is the first test time, and the ordinate is the first signal parameter, and the first slope of each straight line is determined; the signal quality of the signal link is tested according to the multiple first slopes of the multiple straight lines. That is, the embodiment of the present application determines the first signal parameter corresponding to the signal link by testing the signal link, and the first signal parameter can include not only link error, but also eye height, eye width, number of link state changes, etc., that is, it is not limited to the signal quality test of the SAS link, and the signal quality test can also be performed on other signal links; according to the first test time and the first signal parameter of the signal quality test, multiple straight lines are fitted, and then the first slope of each straight line is determined, and the signal quality of the signal link can be predicted according to the first slope. Through the embodiments of the present application, the problem of incomplete coverage of high-speed signal link monitoring scenarios caused by only periodic bit error monitoring of SAS links in the related technology and no monitoring of other high-speed signals can be solved, thereby covering the high-speed signal link monitoring scenarios.

[0150] In an exemplary embodiment, the test module 62 is also used to obtain the business volume of the server cluster where the server corresponding to the signal link is located, and determine the second test time of the last signal quality test on the signal link; calculate the time interval between the second test time and the current time; when it is determined that the business volume is less than or equal to the business volume threshold, and the time interval is greater than or equal to the time interval threshold, determine that the signal quality test of the signal link is allowed; when it is determined that the business volume is greater than the business volume threshold, and / or the time interval is less than the time interval threshold, determine that the signal quality test of the signal link is not allowed.

[0151] In an exemplary embodiment, the test module 62 is also used to control the first test platform to perform a signal quality test on the signal link when it is determined that the signal link is a high-speed serial signal, so as to obtain the eye height corresponding to the signal link and the eye width corresponding to the signal link; and to control the second test platform to perform a signal quality test on the signal link when it is determined that the signal link is a storage link signal, so as to obtain the link error corresponding to the signal link and the number of link state changes corresponding to the signal link.

[0152] In an exemplary embodiment, the prediction module 64 is further used to determine a first size relationship between each first slope and a preset threshold; when the first size relationship indicates that multiple first slopes are greater than or equal to the preset threshold, determine that the signal quality of the signal link has no degradation trend; when the first size relationship indicates that a second slope among the multiple first slopes is greater than or equal to the preset threshold, and the other slopes among the multiple first slopes are less than the preset threshold, determine a first time period corresponding to the second slope and a second time period corresponding to the other slopes, wherein the first time period includes: a first test time corresponding to a data point in a straight line corresponding to the second slope, and the second time period includes: a first test time corresponding to a data point in a straight line corresponding to the other slope, and the other slope is a slope among the multiple first slopes except the second slope; determine a first time point in the first time period and a second time point in the second time period A second size relationship between the first time point and the second time point, wherein the first time point is a time point corresponding to any data point in the first time period except the time point corresponding to the last data point in the straight line corresponding to the second slope, and the second time point is a time point corresponding to any data point in the straight lines corresponding to the other slopes; when the second size relationship indicates that the first time point is greater than the second time point, a second signal quality test is performed on the signal link to determine a second signal parameter of the signal link, and a second prediction is performed on the signal quality of the signal link according to the second signal parameter to determine the signal quality of the signal link according to the first prediction result; when the second size relationship indicates that the first time point is greater than the second time point, it is determined that the signal quality of the signal link has no deterioration trend; when the first size relationship indicates that the multiple first slopes are all less than the preset threshold, it is determined that the signal quality of the signal link has a deterioration trend.

[0153] In an exemplary embodiment, the prediction module 64 is also used to determine whether the first signal parameter is greater than or equal to a signal parameter threshold when it is determined that the signal quality of the signal link has a deterioration trend; when it is determined that the first signal parameter is less than the signal parameter threshold, start the signal link alarm mechanism and record the number of alarms of the signal link in a third time period; when the number of alarms in the third time period is less than or equal to the alarm number threshold, determine that the signal quality of the signal link is updated from having a deterioration trend to having no deterioration trend; when the number of alarms in the third time period is greater than the alarm number threshold, update the preset time interval so that the updated preset time interval is less than the preset time interval; perform a third signal quality test on the signal link based on the updated preset time interval to determine the third signal parameter of the signal link, and perform a third prediction on the signal quality of the signal link based on the third signal parameter to determine the signal quality of the signal link according to the second prediction result.

[0154] In an exemplary embodiment, the prediction module 64 is also used to perform fault elimination processing on the signal link when it is determined that the first signal parameter is greater than or equal to the signal parameter threshold; and to update the signal quality of the signal link from having a degradation trend to having no degradation trend when it is determined that the fault corresponding to the signal link has been eliminated.

[0155] In an exemplary embodiment, the prediction module 64 is also used to perform troubleshooting on the signal link in one of the following ways: by restarting the physical layer interface corresponding to the signal link to troubleshoot the signal link; by isolating the server corresponding to the signal link from the server cluster where the server is located, and restarting the isolated server to troubleshoot the signal link; by reducing the corresponding operating rate of the server to troubleshoot the signal link; by reducing the data transmission bandwidth corresponding to the signal link to troubleshoot the signal link.

[0156] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps in the above method embodiment when running.

[0157] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0158] Optionally, in this embodiment, the storage medium may be configured to store program codes for executing the following steps:

[0159] S1, performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link;

[0160] S2, fitting a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter;

[0161] S3: predicting the signal quality of the signal link according to the multiple first slopes.

[0162] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in the above method embodiment.

[0163] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0164] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0165] S1, performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link;

[0166] S2, fitting a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter;

[0167] S3: predicting the signal quality of the signal link according to the multiple first slopes.

[0168] An embodiment of the present application further provides a computer program product, including a computer program, which is implemented to perform the steps in the above method embodiment when executed by a processor.

[0169] Optionally, in this embodiment, the computer program product may be executed by a processor to perform the following steps:

[0170] S1, performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link;

[0171] S2, fitting a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter;

[0172] S3: predicting the signal quality of the signal link according to the multiple first slopes.

[0173] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0174] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0175] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting signal quality, characterized in that: include: Performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link bit error corresponding to the signal link, and a number of link state changes corresponding to the signal link; fitting multiple straight lines according to a first test time of performing a signal quality test on the signal link and the first signal parameter, and determining a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter; The signal quality of the signal link is predicted according to a plurality of first slopes.

2. The method according to claim 1, characterized in that Before performing a signal quality test on the signal link based on a preset time interval, the method further includes: Acquire the service volume of the server cluster where the server corresponding to the signal link is located, and determine the second test time of the last signal quality test performed on the signal link; Calculating the time interval between the second test time and the current time; In a case where it is determined that the traffic volume is less than or equal to the traffic volume threshold and the time interval is greater than or equal to the time interval threshold, determining to allow a signal quality test to be performed on the signal link; When it is determined that the traffic volume is greater than the traffic volume threshold and / or the time interval is less than the time interval threshold, it is determined that the signal quality test on the signal link is not allowed.

3. The method according to claim 1, characterized in that Performing a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link includes: When it is determined that the signal link is a high-speed serial signal, controlling the first test platform to perform a signal quality test on the signal link to obtain an eye height corresponding to the signal link and an eye width corresponding to the signal link; When it is determined that the signal link is a storage link signal, the second test platform is controlled to perform a signal quality test on the signal link to obtain a link error corresponding to the signal link and a number of link state changes corresponding to the signal link.

4. The method according to claim 1, characterized in that: Predicting the signal quality of the signal link according to the multiple first slopes includes: Determine a first magnitude relationship between each first slope and a preset threshold; When the first magnitude relationship indicates that the plurality of first slopes are all greater than or equal to the preset threshold, determining that the signal quality of the signal link has no degradation trend; In a case where the first magnitude relationship indicates that a second slope among the plurality of first slopes is greater than or equal to the preset threshold, and other slopes among the plurality of first slopes are less than the preset threshold, determining a first time period corresponding to the second slope and a second time period corresponding to the other slopes, wherein the first time period includes: a first test time corresponding to a data point in a straight line corresponding to the second slope, the second time period includes: a first test time corresponding to a data point in a straight line corresponding to the other slopes, and the other slopes are slopes among the plurality of first slopes except the second slope; Determine a second magnitude relationship between a first time point in the first time period and a second time point in the second time period, wherein the first time point is a time point corresponding to any data point in the first time period except a time point corresponding to the last data point in the straight line corresponding to the second slope, and the second time point is a time point corresponding to any data point in the straight lines corresponding to the other slopes; When the second size relationship indicates that the first time point is greater than the second time point, performing a second signal quality test on the signal link to determine a second signal parameter of the signal link, and performing a second prediction on the signal quality of the signal link according to the second signal parameter to determine the signal quality of the signal link according to the first prediction result; When the second magnitude relationship indicates that the first time point is greater than the second time point, determining that the signal quality of the signal link has no degradation trend; When the first magnitude relationship indicates that the plurality of first slopes are all smaller than the preset threshold, it is determined that the signal quality of the signal link has a deterioration trend.

5. The method according to claim 4, characterized in that After predicting the signal quality of the signal link according to the multiple first slopes, the method further includes: In a case where it is determined that the signal quality of the signal link has a deterioration trend, determining whether the first signal parameter is greater than or equal to a signal parameter threshold; When it is determined that the first signal parameter is less than the signal parameter threshold, a signal link alarm mechanism is started, and the number of alarms of the signal link in a third time period is recorded; When the number of alarms in the third time period is less than or equal to the alarm number threshold, determining that the signal quality of the signal link is updated from having a degradation trend to having no degradation trend; When the number of alarms in the third time period is greater than the alarm number threshold, updating the preset time interval so that the updated preset time interval is less than the preset time interval; A third signal quality test is performed on the signal link based on the updated preset time interval to determine a third signal parameter of the signal link, and a third prediction is performed on the signal quality of the signal link based on the third signal parameter to determine the signal quality of the signal link based on the second prediction result.

6. The method according to claim 5, characterized in that After determining whether the first signal parameter is greater than or equal to a signal parameter threshold, the method further includes: In the case where it is determined that the first signal parameter is greater than or equal to the signal parameter threshold, performing fault elimination processing on the signal link; When it is determined that the fault corresponding to the signal link has been eliminated, the signal quality of the signal link is updated from having a degradation trend to having no degradation trend.

7. The method according to claim 6, characterized in that Performing fault elimination processing on the signal link includes: Troubleshoot the signal link by one of the following methods: Restarting the physical layer interface corresponding to the signal link to eliminate the fault of the signal link; The signal link is troubleshooted by isolating the server corresponding to the signal link from the server cluster where the server is located, and restarting the isolated server; By reducing the corresponding operating rate of the server, the signal link is troubleshooted; The signal link is subjected to fault elimination processing by reducing the data transmission bandwidth corresponding to the signal link.

8. A signal quality prediction device, characterized in that: include: A test module, configured to perform a signal quality test on a signal link based on a preset time interval to determine a first signal parameter of the signal link, wherein the first signal parameter includes at least one of the following: an eye height corresponding to the signal link, an eye width corresponding to the signal link, a link error corresponding to the signal link, and a number of link state changes corresponding to the signal link; a fitting module, configured to fit a plurality of straight lines according to a first test time for performing a signal quality test on the signal link and the first signal parameter, and determine a first slope corresponding to each straight line, wherein the abscissa of a test point in each straight line is the first test time, and the ordinate is the first signal parameter; The prediction module is used to predict the signal quality of the signal link according to multiple first slopes.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method according to any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images