Fault detection method and device, equipment and storage medium

Abstract

The invention discloses a fault detection method and device, equipment and a storage medium, and relates to the technical field of servers, in the fault detection method, firstly, current stress pressure distribution data of each contact of a golden finger is collected, and whether a communication fault occurs between a memory bank and a memory slot or not is determined based on the current stress pressure distribution data, so that the fault detection efficiency is improved; repeated plugging and unplugging of the memory bank can be avoided, the fault detection efficiency is improved, and the stability of the memory bank is improved. And secondly, because a contact surface exists between the contact of the golden finger and the memory slot and the current stress pressure distribution data comprises the current stress pressure at a plurality of positions of the contact surface, the current stress pressure at the plurality of positions of the same contact surface can be comprehensively evaluated based on the first fault detection model to obtain a fault detection result. Therefore, the detection accuracy can be improved, and the stability of the memory bank is further improved. Therefore, the problem of poor stability of the memory bank in some technologies is solved.

Description

Technical Field

[0001] The present application relates to the field of server technology, and in particular to a fault detection method, apparatus, device, and storage medium. Background Art

[0002] In servers, memory modules are critical, ultra-high-speed storage media, enabling rapid data read and write operations. They are primarily used for temporary storage of running programs and data, and provide real-time computing and operational support for components like the CPU (Central Processing Unit). As a core component in servers, the stability of the memory module impacts overall server performance.

[0003] Currently, when a memory module fails, maintenance personnel often manually remove and reinstall it to determine if the problem lies with the communication between the module and the memory slot. This method of locating the failure is inefficient, resulting in poor memory module stability and impacting server performance. Summary of the Invention

[0004] The present application provides a fault detection method, a fault detection device, an electronic device, a computer-readable storage medium, and a computer program product to at least solve the problem of poor stability of memory bars in the related art.

[0005] This application provides a fault detection method, including:

[0006] When a memory stick is inserted into a memory slot, current stress distribution data of each contact point of a gold finger is obtained, wherein the memory stick communicates with the memory slot via the contact points of the gold finger, and a contact surface exists between the contact points and the memory slot, and the current stress distribution data represents current stress at multiple positions of the contact surface;

[0007] The current stress distribution data is input into a first fault detection model to obtain a fault detection result, wherein the fault detection result indicates whether a communication fault occurs between the memory module and the memory slot.

[0008] The present application also provides a fault detection device, comprising:

[0009] a data acquisition module, configured to obtain, when a memory stick is inserted into a memory slot, current stress distribution data of each contact point of a gold finger, wherein the memory stick communicates with the memory slot via the contact points of the gold finger, and a contact surface exists between the contact points and the memory slot, and wherein the current stress distribution data represents current stress at multiple locations on the contact surface;

[0010] The fault detection module is used to input the current stress distribution data into a first fault detection model to obtain a fault detection result, wherein the fault detection result indicates whether a communication fault occurs between the memory bar and the memory slot.

[0011] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned fault detection methods when executing the computer program.

[0012] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned fault detection methods are implemented.

[0013] In the technical solutions of some embodiments of the present application, first, by collecting the current stress distribution data of each contact point of the gold finger, and based on the current stress distribution data, determining whether a communication failure has occurred between the memory stick and the memory slot, this can avoid multiple plugging and unplugging of the memory stick, greatly improving the efficiency of fault detection, and thus improving the stability of the memory stick and ensuring server performance. Secondly, because there is a contact surface between the contact point of the gold finger and the memory slot, and the current stress distribution data includes the current stress at multiple locations on the contact surface, it is possible to comprehensively evaluate the current stress at multiple locations on the same contact surface based on the first fault detection model to obtain a fault detection result. In this way, the detection accuracy can be improved, thereby further improving the stability of the memory stick and ensuring server performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0015] Figure 1 This is a diagram showing the relationship between the memory module and the memory slot;

[0016] Figure 2 A flowchart of a fault detection method provided in some embodiments of the present application;

[0017] Figure 3 A schematic diagram of collecting current stress distribution data provided in some embodiments of the present application;

[0018] Figure 4 A schematic diagram of a module of a fault detection system provided in some embodiments of the present application;

[0019] Figure 5A schematic diagram of a module of a fault detection device provided in some embodiments of the present application;

[0020] Figure 6 A schematic diagram of a module of an electronic device provided for some embodiments of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0022] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or precedence.

[0023] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0024] In order to explain this application solution more clearly, the relationship between the memory stick, gold fingers, contacts and memory slots is first explained. Figure 1 , is a schematic diagram of the relationship between the memory stick 11 and the memory slot 12. Figure 1 In the embodiment, the memory slot 12 is usually provided on the server motherboard. The server motherboard may also include components such as a CPU. The memory slot 12 is communicatively connected to components such as the CPU. The memory stick 11 may be a circuit board assembly of a specified shape (such as a rectangle). In the memory stick 11, electronic components such as memory integrated circuits and capacitors for temporarily storing data may be encapsulated. At one end of the memory stick 11, a gold finger 111 may be provided. The gold finger 111 may be a row of conductive sheets, including one or more contacts 1111. When the memory stick 11 is inserted into the memory slot 12, the elastic contact sheet inside the memory slot 12 will apply a certain stress to the contact 1111 of the gold finger 111, so that the contact 1111 is in close contact with the memory slot 12. In this way, an electrical connection can be established between the memory stick 11 and the memory slot 12, and components such as the CPU can also read and write data in the memory stick 11 through the memory slot 12.

[0025] When the memory stick 11 is abnormal (for example, the CPU fails to read or write data from the memory stick), it is usually necessary to first check whether it is a communication failure between the memory stick 11 and the memory slot 12. Among them, the communication failure between the memory stick 11 and the memory slot 12 is mainly manifested in insufficient pressure between the contact 1111 of the gold finger 111 and the memory slot 12. The so-called pressure refers to the pressure per unit area in the contact surface between the contact 1111 and the memory slot 12. The direction of the pressure is perpendicular to the contact surface. The reasons for insufficient pressure may include but are not limited to wear of the contact 1111, design defects of the memory slot 12, and the memory stick 11 not being fully inserted into the memory slot 12.

[0026] At present, in some technologies, when the memory stick is abnormal, the maintenance personnel will re-plug the memory stick 11. If the communication between the memory stick 11 and the memory slot 12 returns to normal after re-plugging, it is considered that the communication fault is resolved. If the communication between the memory stick 11 and the memory slot 12 does not return to normal after re-plugging, it will continue to determine whether the cause is the wear of the contacts 1111, the design defects of the memory slot 12, etc. In this fault location method, on the one hand, the efficiency is low, resulting in poor stability of the memory stick 11, which in turn affects the performance of the server. On the other hand, the memory stick 11 needs to be plugged in and out multiple times. In the process of multiple plugging and unplugging, new fault problems may be introduced (such as the memory stick 11 is damaged), thereby increasing the difficulty of locating the fault problem.

[0027] In other technologies, a first-type pressure sensor senses the pressure between the contact 1111 of the gold finger 111 and the memory slot 12 and outputs a corresponding electrical signal. Specifically, when the pressure between the contact 1111 and the memory slot 12 is less than a threshold value, the first-type pressure sensor can output a first electrical signal (e.g., a high level), indicating that the pressure between the memory stick 11 and the memory slot 12 is insufficient, resulting in a communication failure. When the pressure between the contact 1111 and the memory slot 12 is greater than the threshold value, the first-type pressure sensor can output a second electrical signal (e.g., a low level), indicating that the pressure between the memory stick 11 and the memory slot 12 is sufficient, resulting in no communication failure. In these technologies, considering that the pressures of different contacts 1111 of the gold finger 111 may vary, a first-type pressure sensor is provided under each contact 1111. Each first-type pressure sensor is configured to sense the pressure of one of the contacts 1111 of the gold finger 111 and output a first electrical signal or a second electrical signal. After comprehensively evaluating the electrical signals output by multiple first-type pressure sensors, it can be determined whether a communication failure has occurred between the memory stick 11 and the memory slot 12. For example, if the electrical signals output by most of the first-type pressure sensors are the second electrical signals, it indicates that there is no communication failure between the memory stick 11 and the memory slot 12; if the electrical signals output by most of the first-type pressure sensors are the first electrical signals, it indicates that a communication failure has occurred between the memory stick 11 and the memory slot 12.

[0028] The aforementioned pressure sensor-based technology suffers from inaccurate detection. For example, there is a contact surface between each contact 1111 of the gold finger 111 and the memory slot 12. Normally, as long as the applied pressure at one or more locations on each contact surface exceeds a threshold value, normal communication between the memory stick 11 and the memory slot 12 is guaranteed. However, the first-type pressure sensor under each contact 1111 can only detect the applied pressure at one location on the contact surface. Therefore, when the electrical signal output by the first-type pressure sensor is the first electrical signal, it does not indicate that a communication failure has occurred between the memory stick 11 and the memory slot 12.

[0029] In view of this, the present application provides a fault detection method, which, on the one hand, can improve the efficiency of fault detection, thereby improving the stability of the memory stick and ensuring server performance. On the other hand, there is no need to repeatedly plug and unplug the memory stick, avoiding the introduction of new fault problems during the fault location process. On the other hand, it can improve detection accuracy. The fault detection method can be applied to a fault detection system, or can be applied to an electronic device that runs a fault detection system. Electronic devices may include but are not limited to tablet computers, laptops, desktop computers, servers, integrated circuit boards, etc. In conjunction with reference Figure 2 , which is a flow chart of a fault detection method provided in some embodiments of the present application. Figure 2In the present invention, the fault detection method includes the following steps:

[0030] Step S201, when the memory stick is inserted into the memory slot, obtain the current stress distribution data of each contact of the gold finger. The memory stick communicates with the memory slot through the contacts of the gold finger. There is a contact surface between the contacts and the memory slot. The current stress distribution data represents the current stress at multiple positions of the contact surface.

[0031] See also Figure 3 , which is a schematic diagram of collecting current stress distribution data provided in some embodiments of the present application. Figure 3 In the figure, the square area represents the gold finger area. Each elliptical area represents the contact surface between the gold finger contact and the memory slot. For example, the elliptical area A represents the contact surface between the gold finger contact a and the memory slot, and the elliptical area B represents the contact surface between the gold finger contact b and the memory slot. The small black dots in each elliptical area can represent different positions of the contact surface. For any contact, the current stress at multiple positions on the contact surface of the contact is collected to obtain the current stress distribution data of the contact. For example, the current stress at each position in the elliptical area A is collected to obtain the current stress distribution data of contact a, and the current stress at each position in the elliptical area B is collected to obtain the current stress distribution data of contact b. That is, each contact has its own corresponding current stress distribution data.

[0032] In this embodiment, the current stress distribution data of each contact point of the gold finger can be collected by the second type of pressure sensor. Different from the first type of pressure sensor in some of the above technologies, the second type of pressure sensor in this embodiment can include multiple channels arranged in an array, each channel can independently sense the stress at its location and output an electrical signal representing the magnitude of the stress. For example, Figure 3 For example. The second type of pressure sensor may include a 3*3 array of channels. After deploying a second type of pressure sensor under each elliptical area, the second type of pressure sensor can sense the stress at the positions of the small black dots in the elliptical area and output corresponding electrical signals. For example, assuming that the current stresses at various positions in the elliptical area A are 0.5Pa, 0.45Pa, 0.48Pa, 0.46Pa, 0.5Pa, 0.43Pa, 0.3Pa, 0.25Pa, and 0.36Pa, respectively, the first channel of the second type of pressure sensor under the elliptical area A can output an electrical signal representing 0.5Pa, the second channel can output an electrical signal representing 0.45Pa, and the third channel can output an electrical signal representing 0.48Pa, and so on. In this way, the current stress distribution data of each contact of the gold finger can be obtained.

[0033] Step S202: input the current stress distribution data into a first fault detection model to obtain a fault detection result, which indicates whether a communication fault occurs between the memory module and the memory slot.

[0034] Compared with some technologies that use a first type of pressure sensor to sense the stress at one position on the contact surface, the present application uses a second type of pressure sensor to sense the stress at multiple positions on the same contact surface, and comprehensively evaluates the stress distribution data at multiple positions to obtain the fault detection results between the memory stick and the memory slot. This can avoid the problem of inaccurate detection in some technologies and greatly improve the detection accuracy. Therefore, it can further improve the stability of the memory stick and ensure server performance.

[0035] In summary, in the technical solutions of some embodiments of the present application, first, by collecting the current stress distribution data of each contact of the gold finger, and based on the current stress distribution data, determining whether a communication failure has occurred between the memory stick and the memory slot, this can avoid multiple plugging and unplugging of the memory stick, greatly improving the fault detection efficiency, and thus improving the stability of the memory stick and ensuring server performance. Secondly, because there is a contact surface between the contact of the gold finger and the memory slot, and the current stress distribution data includes the current stress at multiple positions of the contact surface, it is possible to comprehensively evaluate the current stress at multiple positions of the same contact surface based on the first fault detection model to obtain a fault detection result. In this way, the detection accuracy can be improved, thereby further improving the stability of the memory stick and ensuring server performance.

[0036] In some embodiments, if the fault detection result indicates a communication failure between the memory stick and the memory slot, the current stress distribution data can be input into a second fault detection model to determine the cause of the communication failure between the memory stick and the memory slot. Specifically, the cause of the communication failure is insufficient stress between the gold finger of the memory stick and the memory slot, such as worn contacts, design flaws in the memory slot, or the memory stick not being fully inserted into the slot. Based on the cause of the communication failure, a determination can be made as to whether to reseat the memory stick. For example, if the communication failure is caused by an incomplete insertion of the memory stick, the memory stick can be pushed further into the slot without removing it. In this case, reseat / replace the memory stick is unnecessary. For another example, if the communication failure is caused by worn contacts, the memory stick may need to be replaced. In this case, the worn memory stick needs to be removed from the slot and reinserted with a replacement, unworn memory stick. In this case, reseat / replace the memory stick is necessary.

[0037] In the above-described embodiment of the present application, the second fault detection model is used to evaluate the cause of the communication failure between the memory module and the memory slot, and based on the cause of the communication failure, a determination is made as to whether to reseat the memory module. This prevents ineffective insertion and removal of the memory module, reduces the number of insertions and removals, and improves fault detection efficiency. Furthermore, this prevents the introduction of new faults during ineffective insertion and removal of the memory module.

[0038] In some embodiments, after the fault detection results indicate a communication failure between the memory module and the memory slot, and the cause of the communication failure between the memory module and the memory slot is determined using the second fault detection model, the cause of the communication failure and the current stress distribution data can be input into the third fault detection model to determine the required fault repair operation. Based on the fault repair operation, the communication failure between the memory module and the memory slot can be accurately resolved, avoiding ineffective repair attempts, improving fault repair efficiency, and further enhancing the stability of the memory module.

[0039] In some embodiments, the first fault detection model, the second fault detection model, and the third fault detection model may be three different models.

[0040] In other embodiments, at least two of the first, second, and third fault detection models can be the same model. That is, at least two of the first, second, and third fault detection models can be integrated into a single model. For example, after integrating the first, second, and third fault detection models into a single model M, and inputting the current stress distribution data into model M, model M can output a fault detection result. If the fault detection result indicates a communication failure between the memory module and the memory slot, model M can also output the cause of the communication failure and the required fault repair actions.

[0041] In some embodiments, the first fault detection model may be a mathematical statistical model or a machine learning model (such as a neural network model).

[0042] In some embodiments, when the first fault detection model is a mathematical statistical model, the first fault detection model can perform statistical analysis on the current stress distribution data of each contact point of the gold finger to obtain a fault detection result. Specifically, for any contact surface, the first fault detection model can calculate the average stress and deviation index of the contact surface. The deviation index represents the degree of deviation between the stress at each location on the contact surface and the average stress. Based on the average stress and deviation index, it can be determined whether a communication failure has occurred between the contact point corresponding to the contact surface and the memory slot. For example, if the average stress is greater than a first threshold and the deviation index is less than a second threshold, it can be determined that no communication failure has occurred between the contact point corresponding to the contact surface and the memory slot. Conversely, if the average stress is less than the first threshold or the deviation index is greater than the second threshold, it can be determined that a communication failure has occurred between the contact point corresponding to the contact surface and the memory slot. If no communication failure has occurred between the memory slot and most or all contacts, it can be determined that no communication failure has occurred between the memory module and the memory slot. If a communication failure has occurred between the memory slot and most or all contacts, it can be determined that a communication failure has occurred between the memory module and the memory slot. It is understood that the mathematical statistical method may include but is not limited to the method described above, and this application does not impose any limitation on this.

[0043] In some embodiments, the fault detection system of the present application can store historical stress distribution data for the gold fingers of one or more memory sticks. When the first fault detection model is a machine learning model, the historical stress distribution data for the gold fingers of one or more memory sticks can be obtained before the current stress distribution data is input into the first fault detection model. The historical stress distribution data can include faulty historical stress distribution data and normal historical stress distribution data. Faulty historical stress distribution data refers to stress distribution data for each contact point of the gold fingers of the memory stick during the memory stick's historical operation when a communication failure occurred between the memory stick and the memory slot. Normal historical stress distribution data refers to stress distribution data for each contact point of the gold fingers of the memory stick during the memory stick's historical operation when no communication failure occurred between the memory stick and the memory slot. After labeling the faulty historical stress distribution data and the normal historical stress distribution data, the labeled faulty historical stress distribution data and the normal historical stress distribution data can be used to train the first fault detection model. By inputting the current stress distribution data into the trained first fault detection model, the aforementioned fault detection results can be obtained.

[0044] In some embodiments, the second fault detection model and the third fault detection model may be machine learning models. The second fault detection model may be trained based on the following method:

[0045] Find the target memory bank and target memory slot where communication failure occurred during historical operation;

[0046] Obtaining historical fault logs of the target memory stick and the target memory slot at the time of the communication failure, and obtaining first historical stress distribution data of each contact point of the gold finger of the target memory stick at the time of the communication failure;

[0047] Based on historical fault logs, determine the cause of the historical communication failure between the target memory module and the target memory slot;

[0048] The first historical response pressure distribution data is annotated using historical communication failure causes, and the annotated first historical response pressure distribution data is used to train a second fault detection model.

[0049] Specifically, the fault detection system of the present application can save the historical operation logs of the target memory bar and the target memory slot, and the historical stress distribution data of the target memory bar. In the historical operation logs saved by the fault detection system, the historical operation logs of the target memory bar and the target memory slot at the time of the communication failure can be filtered out as the historical fault logs. At the same time, in the historical stress distribution data saved by the fault detection system, the historical stress distribution data of each contact of the gold finger of the target memory bar at the time of the communication failure can be filtered out as the first historical stress distribution data. For example, if a communication failure occurs between the target memory bar and the target memory slot between 10:01 and 10:30, the historical operation logs of the target memory bar and the target memory slot between 10:01 and 10:30 can be filtered out as the historical fault log, and the historical stress distribution data of the gold finger of the target memory bar between 10:01 and 10:30 can be filtered out as the first historical stress distribution data.

[0050] In the historical fault logs of the target memory stick and the target memory slot, the cause of the communication failure between the target memory stick and the target memory slot can be extracted. Based on the extracted communication failure cause, the first historical stress distribution data can be annotated. For example, assuming:

[0051] Based on the historical fault logs of memory stick T1 and memory slot C1 from 10:01 to 10:30, the communication fault cause Y1 is extracted, and the first historical stress distribution data D1 of the gold finger of memory stick T1 from 10:01 to 10:30 is obtained;

[0052] Based on the historical fault logs of memory stick T2 and memory slot C2 from 16:00 to 16:10, the communication fault cause Y2 is extracted, and the first historical stress distribution data D2 of the gold finger of memory stick T2 from 16:00 to 16:10 is obtained;

[0053] Based on the historical fault logs of the memory stick T1 and the memory slot C1 from 18:03 to 18:10, the communication fault cause Y3 is extracted, and the first historical stress distribution data D3 of the gold finger of the memory stick T1 from 18:03 to 18:10 is obtained.

[0054] The first historical pressure distribution data D1 can be annotated with the communication fault cause Y1, the first historical pressure distribution data D2 can be annotated with the communication fault cause Y2, and the first historical pressure distribution data D3 can be annotated with the communication fault cause Y3. Based on the annotated first historical pressure distribution data D1, D2, and D3, a second fault detection model can be trained.

[0055] In the above embodiment, since the historical fault log can more accurately reflect the historical operating conditions of the target memory bar and the target memory slot, the communication fault cause extracted from the historical fault log is more accurate, thereby ensuring the accuracy of the model training.

[0056] In some embodiments, after the communication fault between the memory module and the memory slot is resolved, the fault detection system may also receive and save a fault repair operation entered by maintenance personnel to resolve the communication fault. In this manner, the first historical stress distribution data may be annotated based on the fault repair operation saved in the fault detection system. The annotated first historical stress distribution data may be used to train the aforementioned third fault detection model.

[0057] In some embodiments, during the use of the memory stick and the memory slot, the stress between the gold finger of the memory stick and the memory slot is a process of gradually decreasing. For example, as the use time of the memory stick or the memory slot gradually increases, the contacts of the gold finger or the memory slot will gradually wear out, causing the stress between the gold finger and the memory slot to gradually decrease. Therefore, when there is no communication failure between the memory stick and the memory slot, the target time point when the communication failure between the memory stick and the memory slot occurs can be estimated based on the stress change trend between the gold finger and the memory slot, and the memory stick or memory slot can be replaced in advance before the target time point. In this way, the communication failure problem between the memory stick and the memory slot can be avoided in advance, thereby greatly improving the stability of the memory stick.

[0058] Based on the above description, in some embodiments, if the fault detection result indicates that there is no communication failure between the memory module and the memory slot, the contact pressure of any contact of the gold finger can be determined based on the current pressure distribution data of the contact point. Based on the pressure difference between the contact pressure and the reference pressure, a first alarm message can be generated, and the first alarm message includes a time point for replacing the memory module or memory slot.

[0059] Specifically, in some embodiments, for any contact point of the gold finger, the current stress distribution data of the contact point can be averaged, and the average calculation result can be used as the contact stress of the contact point. For example, the current stress distribution data of contact a1 is averaged to obtain the contact stress F1 of contact a1. The current stress distribution data of contact a2 is averaged to obtain the contact stress F2 of contact a2. And so on.

[0060] In other embodiments, for any contact of the gold finger, the maximum stress can be selected from the current stress distribution data of that contact and used as the contact stress of that contact. For example, if the maximum stress in the current stress distribution data of contact a3 is 0.53 Pa, then 0.53 Pa can be used as the contact stress F3 of contact a3. Similarly, if the maximum stress in the current stress distribution data of contact a4 is 0.45 Pa, then 0.45 Pa can be used as the contact stress F4 of contact a4, and so on.

[0061] In other embodiments, for any contact point of the gold finger, the minimum stress can be screened out from the current stress distribution data of the contact point and used as the contact stress of the contact point.

[0062] It should be noted that the calculation logic of the contact pressure may include but is not limited to the logic listed above, and this application does not impose any restrictions on this.

[0063] The contact pressures of different contacts can be the same or different. After obtaining the contact pressures of each contact of the gold finger, the contact pressures of multiple contacts of the gold finger can be fused and calculated (e.g., averaged), and the result of the fused calculation is used as the target contact pressure for subsequent calculations.

[0064] In this embodiment, if the gold finger has multiple different contact pressures, the contact pressure with the minimum value is found among the multiple different contact pressures and used as the target contact pressure. The first alarm information can be generated based on the pressure difference between the target contact pressure and the reference pressure.

[0065] Specifically, the target contact pressure can be subtracted from the reference pressure to obtain the pressure difference between the target contact pressure and the reference pressure. The pressure difference between the target contact pressure and the reference pressure can be used to estimate the time point for replacing the memory stick or memory slot. For example:

[0066] When the pressure difference between the target contact pressure and the reference pressure is 10% to 15% of the reference pressure, it can be considered that the probability of a communication failure between the memory module and the memory slot in the first preset time period in the future (for example, the next six months) is relatively low. Therefore, the first alarm message can prompt the user to replace the memory module and the memory slot according to the normal maintenance plan.

[0067] When the pressure difference between the target contact pressure and the reference pressure is 15% to 20% of the reference pressure, it can be considered that there is a high probability that a communication failure will occur between the memory module and the memory slot within a second preset time period (e.g., within the next 48 hours). Therefore, the first alarm message may prompt that the memory module or memory slot needs to be replaced in a timely manner within the next 48 hours.

[0068] When the pressure difference between the target contact response pressure and the reference response pressure is more than 20% of the reference response pressure, it can be considered that the memory stick and the memory slot may have a communication failure at any time. Therefore, the first alarm message can prompt that the memory stick or memory slot needs to be replaced immediately to avoid data loss.

[0069] In some embodiments, based on the pressure difference range of the pressure difference between the target contact response pressure and the reference response pressure, the urgency of the first alarm message can be identified based on font color, font size, etc. For example, when the pressure difference between the target contact response pressure and the reference response pressure is 10% to 15% of the reference response pressure, the font color of the first alarm message can be black; when the pressure difference between the target contact response pressure and the reference response pressure is 15% to 20% of the reference response pressure, the font color of the first alarm message can be blue; when the pressure difference between the target contact response pressure and the reference response pressure is more than 20% of the reference response pressure, the font color of the first alarm message can be red. In this way, maintenance personnel can distinguish the urgency of replacing the memory stick or memory slot based on the font color of the first alarm message.

[0070] In some embodiments, the above-mentioned reference stress can be obtained based on the following method:

[0071] Find the target memory bank and target memory slot where communication failure occurred during historical operation;

[0072] Second historical stress distribution data of each contact point of the gold finger of the target memory stick at the communication failure time point is obtained, and a reference stress is obtained based on the second historical stress distribution data.

[0073] Specifically, based on the second historical stress distribution data, the target stress of each contact point of the gold finger when a communication failure occurs can be determined, and then the stress greater than the target stress can be used as the reference stress. The reference stress determined in this way is relatively accurate.

[0074] In some embodiments, when the fault detection result indicates that there is no communication failure between the memory stick and the memory slot, the target time point at which a communication failure will occur between the memory stick and the memory slot in the future can also be predicted based on the current stress distribution data. Based on the time difference between the target time point and the current time point, a second alarm message is generated, and the second alarm message includes the replacement time point of the memory stick or memory slot. Specifically, the current stress distribution data can be input into a trained prediction model, and the prediction model outputs the target time point. Based on the time difference between the target time point and the current time point, the replacement time point of the memory stick or memory slot can be estimated. For example:

[0075] If the time difference between the target time point and the current time point is greater than 180 days, the second alarm message may prompt that the memory module and memory slot should be replaced according to the normal maintenance plan;

[0076] If the time difference between the target time point and the current time point is less than 48 hours, the second alarm message may prompt that the memory stick or memory slot needs to be replaced in time within the next 48 hours;

[0077] If the time difference between the target time point and the current time point is less than 24 hours, the second alarm information may prompt that the memory stick or memory slot needs to be replaced immediately.

[0078] In some embodiments, after obtaining the current stress distribution data, the current stress distribution data of multiple contacts can be aggregated and statistically analyzed to obtain and display a stress distribution map of the gold finger. This allows maintenance personnel to intuitively view the stress levels of various areas of the gold finger.

[0079] Based on the above description, refer to Figure 4 , which is a module diagram of a fault detection system provided in some embodiments of the present application. Figure 4In the embodiment, the fault detection system includes a stress collection module 401, a stress analysis module 402, and an alarm module 403. The stress collection module 401 can receive the electrical signal output by the second-type pressure sensor to obtain current stress distribution data for each contact of the gold finger. The stress analysis module 402 can analyze the current stress distribution data to obtain a fault detection result. If the fault detection result indicates a communication failure between the memory module and the memory slot, the stress analysis module 402 can also output the cause of the communication failure, fault repair operations, etc. If the fault detection result indicates no communication failure between the memory module and the memory slot, the stress analysis module 402 can also determine the contact stress of the contact based on the current stress distribution data of the contact, and output the pressure difference between the contact stress and the reference stress to the alarm module 403. Alternatively, the stress analysis module 402 can also predict the target time point of a future communication failure between the memory module and the memory slot based on the current stress distribution data, and output the time difference to the alarm module 403. The alarm module 403 may determine an alarm level based on the received pressure difference or time difference and generate alarm information.

[0080] 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 the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0081] See also Figure 5 , which is a module schematic diagram of a fault detection device provided in some embodiments of the present application. Figure 5 In the embodiment, the fault detection device comprises:

[0082] The data acquisition module 501 is configured to obtain current stress distribution data of each contact point of a gold finger when a memory stick is inserted into a memory slot. The memory stick communicates with the memory slot via the gold finger contacts, and a contact surface exists between the contacts and the memory slot. The current stress distribution data represents the current stress at multiple locations on the contact surface.

[0083] The fault detection module 502 is configured to input the current stress distribution data into a first fault detection model to obtain a fault detection result, which indicates whether a communication fault occurs between the memory module and the memory slot.

[0084] In some embodiments, if the fault detection result indicates that a communication fault occurs between the memory module and the memory slot, the fault detection module 502 is further configured to:

[0085] Inputting the current stress distribution data into a second fault detection model to obtain a cause of a communication failure between the memory module and the memory slot;

[0086] Based on the cause of the communication failure, determine whether to reseat the memory module.

[0087] In some embodiments, the fault detection module 502 trains the second fault detection model based on the following method:

[0088] Find the target memory bank and target memory slot where communication failure occurred during historical operation;

[0089] Obtaining historical fault logs of the target memory stick and the target memory slot at the time of the communication failure, and obtaining first historical stress distribution data of each contact point of the gold finger of the target memory stick at the time of the communication failure;

[0090] Based on historical fault logs, determine the cause of the historical communication failure between the target memory module and the target memory slot;

[0091] The first historical response pressure distribution data is annotated using historical communication failure causes, and the annotated first historical response pressure distribution data is used to train a second fault detection model.

[0092] In some embodiments, if the fault detection result indicates that there is no communication fault between the memory module and the memory slot, the fault detection module 502 is further configured to:

[0093] For any contact point of the gold finger, the contact pressure of the contact point is determined based on the current pressure distribution data of the contact point;

[0094] Based on the pressure difference between the contact pressure and the reference pressure, a first warning message is generated. The first warning message includes a replacement time point of the memory module or memory slot. The reference pressure is obtained based on the following method:

[0095] Find the target memory bank and target memory slot where communication failure occurred during historical operation;

[0096] Second historical stress distribution data of each contact point of the gold finger of the target memory stick at the communication failure time point is obtained, and a reference stress is obtained based on the second historical stress distribution data.

[0097] In some embodiments, the fault detection module 502 is specifically configured to:

[0098] If there are multiple different contact pressures for the gold finger, then find the contact pressure with the minimum value among the multiple different contact pressures and use it as the target contact pressure;

[0099] First warning information is generated based on a pressure difference between the target contact response pressure and the reference response pressure.

[0100] In some embodiments, after obtaining the current stress distribution data, the fault detection module 502 is further configured to:

[0101] The current stress distribution data of multiple contacts are summarized and statistically analyzed to obtain and display the stress distribution diagram of the gold finger.

[0102] In some embodiments, when the fault detection result indicates that no communication fault occurs between the memory module and the memory slot, the fault detection module 502 is further configured to:

[0103] Based on the current stress distribution data, predict the target time point when communication failure between the memory module and the memory slot will occur in the future;

[0104] Based on the time difference between the target time point and the current time point, a second alarm message is generated, where the second alarm message includes a replacement time point of the memory stick or the memory slot.

[0105] For the description of the features in the embodiment corresponding to the fault detection device, reference can be made to the relevant description of the embodiment corresponding to the sample data processing method, which will not be repeated here.

[0106] See also Figure 6 An embodiment of the present application further provides an electronic device, comprising a memory 10 and a processor 20, wherein the memory 10 stores a computer program, and the processor 20 is configured to run the computer program to execute the steps in any one of the above-mentioned fault detection method embodiments.

[0107] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned fault detection method embodiments when running.

[0108] 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.

[0109] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above-mentioned fault detection method embodiments are implemented.

[0110] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above-mentioned fault detection method embodiments are implemented.

[0111] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0112] The above is a detailed introduction to a fault detection method, device, equipment and storage medium provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

Claims

1. A fault detection method, characterized in that: The method comprises: When a memory stick is inserted into a memory slot, current stress distribution data of each contact point of a gold finger is obtained, wherein the memory stick communicates with the memory slot via the contact points of the gold finger, and a contact surface exists between the contact points and the memory slot, and the current stress distribution data represents current stress at multiple positions of the contact surface; The current stress distribution data is input into a first fault detection model to obtain a fault detection result, wherein the fault detection result indicates whether a communication fault occurs between the memory module and the memory slot.

2. The method according to claim 1, characterized in that If the fault detection result indicates that a communication fault occurs between the memory module and the memory slot, the method further includes: Inputting the current stress distribution data into a second fault detection model to obtain a cause of a communication failure between the memory module and the memory slot; Based on the cause of the communication failure, determine whether to perform an operation of re-inserting or removing the memory module.

3. The method according to claim 2, characterized in that The second fault detection model is trained based on the following method: Find the target memory bank and target memory slot where communication failure occurred during historical operation; Obtaining historical fault logs of a target memory stick and a target memory slot at a communication failure time point, and obtaining first historical stress distribution data of each contact point of a gold finger of the target memory stick at a communication failure time point; Determining a cause of a historical communication failure between a target memory bank and a target memory slot based on the historical failure log; The first historical communication fault causes are used to annotate the first historical stress distribution data, and the annotated first historical stress distribution data are used to train the second fault detection model.

4. The method according to claim 2, characterized in that If the fault detection result indicates that no communication fault occurs between the memory module and the memory slot, the method further includes: For any contact point of the golden finger, determining the contact pressure of the contact point based on the current pressure distribution data of the contact point; Based on a pressure difference between the contact response pressure and a reference response pressure, first warning information is generated, the first warning information including a replacement time point of the memory module or the memory slot, and the reference response pressure is obtained based on the following method: Find the target memory bank and target memory slot where communication failure occurred during historical operation; Second historical stress distribution data of each contact point of the gold finger of the target memory stick at the communication failure time point is obtained, and the reference stress is obtained based on the second historical stress distribution data.

5. The method according to claim 4, characterized in that The generating of the first warning information based on the pressure difference between the contact response pressure and the reference response pressure includes: If the gold finger has multiple different contact pressures, then find the contact pressure with the minimum value among the multiple different contact pressures and use it as the target contact pressure; The first warning information is generated based on a pressure difference between the target contact response pressure and the reference response pressure.

6. The method according to claim 1 or 2, characterized in that After obtaining the current stress distribution data, the method further includes: The current stress distribution data of the multiple contacts are summarized, and the summarized current stress distribution data are statistically analyzed to obtain and display the stress distribution diagram of the gold finger.

7. The method according to claim 1, characterized in that When the fault detection result indicates that no communication fault occurs between the memory module and the memory slot, the method further includes: Predicting a target time point at which a communication failure between the memory module and the memory slot will occur in the future based on the current stress distribution data; Based on the time difference between the target time point and the current time point, second alarm information is generated, where the second alarm information includes a replacement time point of the memory bar or the memory slot.

8. A fault detection device, characterized in that: The device comprises: a data acquisition module, configured to obtain, when a memory stick is inserted into a memory slot, current stress distribution data of each contact point of a gold finger, wherein the memory stick communicates with the memory slot via the contact points of the gold finger, and a contact surface exists between the contact points and the memory slot, and wherein the current stress distribution data represents current stress at multiple locations on the contact surface; The fault detection module is used to input the current stress distribution data into a first fault detection model to obtain a fault detection result, wherein the fault detection result indicates whether a communication fault occurs between the memory bar and the memory slot.

9. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the fault detection method according to any one of claims 1 to 7 when executing the computer program.

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

Images