A method and system for solid state hard disk fault detection based on electrical variables
Through the fault detection method based on electrical variables, the operating parameters abnormalities of the solid-state drive and the task queue are reordered, which solves the problems of solid-state drive failure prediction and early response, and achieves data security and extension of the solid-state drive life.
Patent Information
- Application Number
- CN202110938140.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-16
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-08-16
AI Technical Summary
The existing technology lacks effective prediction and early response measures for solid-state drive failures, which leads to the inability to prevent solid-state drives in advance when they fail, which may lead to the loss of important data.
Using fault detection method based on electrical variables, the training model predicts the abnormal operating parameters of the solid-state drive, and reorders the task queues based on the prediction results, ensuring priority execution of emergency tasks, avoiding data losses, and identifying the fault location and cause, and conducting timely Maintenance or scrap.
It realizes effective prediction and early response to solid-state drive failures, ensures that important tasks are completed before the failure occurs, avoid data loss, and can identify the location and cause of the fault, carry out timely maintenance or scrapping, and extend the service life of the solid-state drive.
Smart Images

Figure CN113590406B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of solid state hard disks, and in particular to a method and system for solid state hard disk fault detection based on electrical variables. Background Art
[0002] With the widespread application of solid-state drives in various fields, in some cases, solid-state drives may malfunction due to the influence of external environmental factors, which may manifest as abnormal conditions such as failure to identify the solid-state drive or loss of internal data in the solid-state drive.
[0003] When a solid-state drive fails, the current response is basically to return the solid-state drive to the original manufacturer for repair, find the fault report by checking the operation log of the solid-state drive's main control chip, and then reproduce the fault phenomenon based on the fault report to locate the fault problem.
[0004] However, these countermeasures are implemented only after the SSD fails, and there is a lack of technical solutions for effectively predicting the failure of the SSD and then responding in advance to the possible failure of the SSD based on the prediction results. Summary of the invention
[0005] The main purpose of the present invention is to provide a method and system for solid-state hard disk fault detection based on electrical variables, aiming to solve the current problem of lack of technical solutions for effectively predicting solid-state hard disk faults and then responding in advance to possible solid-state hard disk faults based on the prediction results.
[0006] The technical solution proposed by the present invention is:
[0007] The present invention proposes a method for solid state hard disk fault detection based on electrical variables, which is applied to a system for solid state hard disk fault detection based on electrical variables; the system comprises a host computer and a detection unit; the detection unit is communicatively connected with the host computer; the method comprises:
[0008] Obtaining a fault prediction model trained by the host computer;
[0009] Detecting the real-time operating parameters of the solid state drive through the detection unit, and determining whether the real-time operating parameters are abnormal;
[0010] When the real-time operating parameters are abnormal, the real-time operating parameters of the solid state drive are substituted as input data into the fault prediction model, and the output data obtained is the predicted fault data;
[0011] The tasks in the task queue of the solid state drive are reordered according to the predicted fault type and the predicted fault occurrence time in the predicted fault data.
[0012] Preferably, the acquiring of the fault prediction model trained by the host computer includes:
[0013] Establishing a fault prediction model through the host computer;
[0014] Perform a fault test to obtain corresponding test fault data after the solid state drive runs for a set time under several groups of test running parameters, wherein the test fault data includes the test fault type and the test fault occurrence time;
[0015] The fault prediction model is trained by using several groups of the test operation parameters as training input data and the corresponding test fault data as training output data.
[0016] Preferably, the operating parameters include reset circuit operating parameters, power supply circuit operating parameters and clock circuit operating parameters; the fault types include reset circuit abnormality, power supply circuit abnormality and clock circuit abnormality; the fault test is performed to obtain corresponding test fault data of the solid state drive after running for a set time under several groups of test operating parameters, including:
[0017] Perform a reset circuit fault test to obtain first test fault data corresponding to the solid state drive after the solid state drive runs for a set time under several sets of reset circuit test operating parameters;
[0018] Perform a power supply circuit fault test to obtain corresponding second test fault data after the solid state drive runs for a set time under several sets of power supply circuit test operating parameters;
[0019] A clock circuit fault test is performed to obtain corresponding third test fault data after the solid state hard disk runs for a set time under several sets of clock circuit test operating parameters.
[0020] Preferably, detecting the real-time operating parameters of the solid state drive by the detection unit and determining whether the real-time operating parameters are abnormal includes:
[0021] Respectively detecting the real-time operating parameters of the reset circuit, the power supply circuit and the clock circuit of the solid state drive;
[0022] respectively judging whether there are abnormalities in the real-time operating parameters of the reset circuit, the real-time operating parameters of the power supply circuit, and the real-time operating parameters of the clock circuit;
[0023] When the real-time operating parameters are abnormal, the real-time operating parameters of the solid state drive are substituted as input data into the fault prediction model, and the output data obtained is the predicted fault data, including:
[0024] When the real-time operating parameters are abnormal, marking an abnormal circuit of the solid state drive;
[0025] The real-time operating parameters of the abnormal circuit of the solid state drive are substituted into the fault prediction model as input data, and the output data obtained is the predicted fault data.
[0026] Preferably, the reordering of the tasks in the task queue of the solid state drive according to the predicted fault type and the predicted fault occurrence time in the predicted fault data includes:
[0027] Obtaining the importance level and original order of each task in the task queue;
[0028] Reordering the tasks in the task queue of the solid state drive according to the predicted fault type, predicted fault occurrence time, importance level and original order of each task, and marking each task with a reordering mark and an original order mark;
[0029] Process each task in the task queue of the solid state drive in the order of the reordering mark.
[0030] Preferably, the processing of each task in the task queue of the solid state drive according to the order of the reordering mark further includes:
[0031] Detecting whether the abnormality of the real-time operating parameter is resolved;
[0032] If so, the remaining tasks in the task queue of the solid state drive are processed in the order of the original sorting marks.
[0033] Preferably, it also includes:
[0034] When the host computer in communication with the solid state drive can identify the solid state drive, determining whether the host computer can access the cache unit of the solid state drive;
[0035] If yes, the host computer generates first fault information, wherein the first fault information is used to indicate that the fault location of the solid state drive is the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit poor soldering and / or circuit short circuit;
[0036] If not, second fault information is generated by the host computer, wherein the second fault information is used to indicate that the fault location of the solid state drive is the cache unit power supply circuit or the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit cold soldering and / or circuit short circuit;
[0037] When the host computer cannot identify the solid-state hard disk, the third fault information is generated by the host computer, wherein the third fault information is used to indicate that the fault location of the solid-state hard disk is the main control chip power supply circuit, and the fault causes of the solid-state hard disk include circuit poor soldering and / or circuit short circuit.
[0038] Preferably, it also includes:
[0039] Acquire the total operating current of the solid state hard disk through the detection unit;
[0040] Sending an erase instruction or a soft destroy instruction to the solid-state hard disk through the host computer that is communicatively connected to the solid-state hard disk, and marking the time when the host computer sends the erase instruction or the soft destroy instruction to the solid-state hard disk as the starting time;
[0041] Determine whether the average value of the total working current in the preset time period after the starting moment is greater than the average value of the total working current in the preset time period before the starting moment;
[0042] If yes, fourth fault information is generated by the host computer, wherein the fourth fault information is used to indicate that the fault location of the solid state drive does not include flash memory particles;
[0043] If not, fifth fault information is generated by the host computer, wherein the fifth fault information is used to indicate that the fault location of the solid state drive is the flash memory particle.
[0044] Preferably, it also includes:
[0045] Removing the flash memory particles of the solid state drive, and establishing a communication connection between the removed flash memory particles and the host computer;
[0046] Get the remaining erase and write times C and the average daily data write volume R of the flash memory particles of the solid-state drive d ;
[0047] The remaining lifespan T of the flash memory particles is calculated based on the remaining erase and write times and the average daily data write amount. s :
[0048]
[0049] Wherein, R is the total capacity of the flash memory particles, in GB; R d The unit is GB, T s The unit of is day;
[0050] When the remaining life is less than a preset life, the flash memory particles are scrapped and the data in the flash memory particles are deleted;
[0051] When the remaining life is not less than the preset life, the flash memory particles are equipped with a normal main control chip and a cache unit.
[0052] The present invention also proposes a system for solid-state hard disk fault detection based on electrical variables. The system applies the method for solid-state hard disk fault detection based on electrical variables as described in any one of the above; the system includes a host computer and a detection unit; the detection unit is communicatively connected to the host computer.
[0053] Through the above technical solution, the following beneficial effects can be achieved:
[0054] In the technical solution proposed in the present invention, the upper computer will reorder the tasks in the task queue of the solid-state hard disk before predicting the time when the fault will occur, so as to give priority to the urgent and important tasks and execute relatively unimportant tasks later. This can ensure that the solid-state hard disk completes the urgent and important tasks before the predicted time when the fault will occur, so as to avoid the loss of important data. In other words, the method for solid-state hard disk fault detection based on electrical variables proposed in the present invention can effectively predict the faults of the solid-state hard disk, and can also respond in advance to possible faults of the solid-state hard disk based on the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.
[0056] Figure 1 This is a flow chart of a first embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention. DETAILED DESCRIPTION
[0057] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.
[0058] The present invention provides a method and system for solid state hard disk fault detection based on electrical variables.
[0059] As attached Figure 1 As shown, in a first embodiment of a method for performing solid-state hard disk fault detection based on electrical variables proposed by the present invention, this embodiment is applied to a system for performing solid-state hard disk fault detection based on electrical variables; the system includes a host computer and a detection unit; the detection unit is communicatively connected to the host computer; this embodiment includes the following steps:
[0060] Step S110: Acquire the fault prediction model trained by the host computer.
[0061] Specifically, a fault prediction model trained by the host computer is obtained. The fault prediction model here is used to predict when a fault will occur during the operation of the solid state drive and the type of the fault.
[0062] Step S120: detecting the real-time operating parameters of the solid state drive by the detection unit, and determining whether the real-time operating parameters are abnormal.
[0063] Specifically, the real-time operating parameters of the solid state drive are detected by the detection unit, and the detected real-time operating parameters are sent to the host computer, and the host computer determines whether the real-time operating parameters are abnormal.
[0064] Specifically, the detection units here may include temperature detection devices, humidity detection devices, vibration detection devices and electric variable detection devices, and the operating parameters may include temperature (detected by temperature detection devices), humidity (detected by humidity detection devices), vibration frequency and vibration amplitude (detected by vibration detection devices), current (detected by current detection tools in electric variable detection devices), and voltage (detected by voltage detection tools in electric variable detection devices).
[0065] Therefore, the solid state hard disk fault detection method of the present invention can detect a variety of abnormal operating parameters such as temperature, humidity, vibration, current and voltage, and has a wide range of applications.
[0066] Specifically in this embodiment, the detection unit is preferably a voltage detection tool, and the real-time operating parameters include operating voltage parameters of different hardware circuits of the solid state drive.
[0067] Step S130: When the real-time operating parameters are abnormal, the real-time operating parameters of the solid state drive are substituted into the fault prediction model as input data, and the obtained output data is the predicted fault data.
[0068] Specifically, when the real-time operating parameters are abnormal, the host computer substitutes the real-time operating parameters of the solid state drive as input data into the fault prediction model, and the output data obtained is the predicted fault data. The predicted fault data here includes the predicted fault type and the predicted fault occurrence time.
[0069] Step S140: reordering the tasks in the task queue of the solid state drive according to the predicted fault type and the predicted fault occurrence time in the predicted fault data.
[0070] Specifically, the host computer reorders the tasks in the task queue of the solid state drive according to the predicted fault type and the predicted fault occurrence time in the predicted fault data.
[0071] For example, if the predicted fault type in the predicted fault data is a power supply circuit voltage fault, and the predicted fault occurrence time is 3 minutes later, then the host computer will reorder the tasks in the task queue of the solid-state hard disk before the predicted fault occurrence time (3 minutes), so as to give priority to urgent and important tasks and execute relatively unimportant tasks later. This can ensure that the solid-state hard disk completes urgent and important tasks before the predicted fault occurrence time to avoid the loss of important data. In other words, the method for solid-state hard disk fault detection based on electrical variables proposed in the present invention can effectively predict the faults of the solid-state hard disk, and can also respond in advance to possible faults of the solid-state hard disk based on the prediction results.
[0072] In a second embodiment of a method for detecting solid state hard disk faults based on electrical variables proposed by the present invention, based on the first embodiment, step S110 includes the following steps:
[0073] Step S210: establishing a fault prediction model through the host computer.
[0074] Specifically, a fault prediction model is established by the host computer.
[0075] Step S220: Perform a fault test to obtain corresponding test fault data after the solid state drive runs for a set time under several groups of test operation parameters, wherein the test fault data includes the test fault type and the test fault occurrence time.
[0076] Specifically, a fault test is performed, and the host computer obtains corresponding test fault data after the solid state drive runs for a set time (for example, 24 hours) under several sets of test operating parameters. The test fault data includes the test fault type and the test fault occurrence time.
[0077] The test operation parameters here are set as constantly changing voltage parameters, that is, when the solid-state hard disk is running under the conditions of the test operation parameters, various test fault data will appear in the solid-state hard disk, which may cause a failure or normal operation. In this way, the corresponding test fault data can include various operating conditions of the solid-state hard disk, so that the test operation parameters and test fault data can be input into the fault prediction model for training, thereby improving the prediction accuracy of the fault prediction model.
[0078] Step S230: using several groups of the test operation parameters as training input data and corresponding test fault data as training output data to train the fault prediction model.
[0079] Specifically, the host computer uses several groups of test operation parameters as training input data and the corresponding test fault data as training output data to train the fault prediction model. This embodiment is a specific solution for establishing a fault prediction model and training the fault prediction model.
[0080] In a third embodiment of a method for detecting faults of a solid-state hard disk based on electrical variables proposed by the present invention, based on the second embodiment, the hardware circuit of the solid-state hard disk mainly includes a reset circuit, a power supply circuit and a start / stop circuit, and the operation of these circuits is directly related to the operation of the entire solid-state hard disk; the operation parameters include reset circuit operation parameters, power supply circuit operation parameters and clock circuit operation parameters; the fault types include reset circuit abnormality, power supply circuit abnormality and clock circuit abnormality; step S220 includes the following steps:
[0081] Step S310: Perform a reset circuit fault test to obtain corresponding first test fault data after the solid state drive runs for a set time under several sets of reset circuit test operating parameters.
[0082] Specifically, a reset circuit fault test is performed on the solid-state hard disk, and the host computer obtains the first test fault data of the solid-state hard disk after it runs for a set time (for example, 24 hours) under several groups of reset circuit test operating parameters (i.e., several groups of different reset circuit test operating parameters, preferably the working voltage of the reset current).
[0083] The test operation parameters here are preferably updated every 30 seconds, and the corresponding first operation result data is synchronously updated every 30 seconds; this embodiment provides a specific solution for collecting model training data for the reset circuit.
[0084] Step S320: Perform a power supply circuit fault test to obtain corresponding second test fault data after the solid state drive runs for a set time under several sets of power supply circuit test operating parameters.
[0085] Specifically, a power supply circuit fault test is performed on the solid-state hard disk, and the host computer obtains the second test fault data of the solid-state hard disk running within a set time (for example, 24 hours) under several groups of power supply circuit test operating parameters (i.e., several groups of different power supply circuit test operating parameters, preferably the working voltage of the power supply circuit).
[0086] The test operation parameters here are preferably updated every 30 seconds, and the corresponding second operation result data is synchronously updated every 30 seconds; this embodiment provides a specific solution for collecting model training data for the power supply circuit.
[0087] Step S330: Perform a clock circuit fault test to obtain corresponding third test fault data after the solid state drive runs for a set time under several sets of clock circuit test operating parameters.
[0088] Specifically, a clock circuit fault test is performed on the solid-state hard disk, and the host computer obtains the third test fault data of the solid-state hard disk running within a set time (for example, 24 hours) under several groups of clock circuit test operating parameters (i.e., several groups of different clock circuit test operating parameters, preferably the working voltage of the clock circuit).
[0089] The test operation parameters here are preferably updated every 30 seconds, and the corresponding third operation result data is synchronously updated every 30 seconds; this embodiment provides a specific solution for collecting model training data for clock circuits.
[0090] In a fourth embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention, based on the third embodiment, step S120 includes the following steps:
[0091] Step S410: respectively detecting the real-time operating parameters of the reset circuit, the power supply circuit and the clock circuit of the solid state drive.
[0092] Specifically, the host computer detects the real-time operating parameters of the reset circuit, the power supply circuit and the clock circuit of the solid state drive respectively.
[0093] Step S420: Determine whether there are any abnormalities in the real-time operating parameters of the reset circuit, the real-time operating parameters of the power supply circuit, and the real-time operating parameters of the clock circuit.
[0094] Specifically, the host computer determines whether there are abnormalities in the real-time operating parameters of the reset circuit, the real-time operating parameters of the power supply circuit, and the real-time operating parameters of the clock circuit.
[0095] The judgment method here is:
[0096] Determine whether the real-time operating parameters of the reset circuit (i.e., the real-time voltage of the reset circuit) fall within the normal voltage range of the reset circuit (e.g., 0.4V-3.3V). If so, it means that the real-time operating parameters of the reset circuit are normal; if not, it means that the real-time operating parameters of the reset circuit are abnormal.
[0097] Determine whether the real-time operating parameters of the power supply circuit (i.e., the real-time voltage of the power supply circuit) fall within the normal voltage range of the power supply circuit (e.g., 1.5V-3.3V). If so, it means that the real-time operating parameters of the power supply circuit are normal; if not, it means that the real-time operating parameters of the power supply circuit are abnormal.
[0098] Determine whether the real-time operating parameters of the clock circuit (i.e., the real-time voltage of the clock circuit) fall within the normal voltage range of the clock circuit (e.g., 1.5V-3.3V). If so, it means that the real-time operating parameters of the clock circuit are normal; if not, it means that the real-time operating parameters of the clock circuit are abnormal.
[0099] Step S130 includes the following steps:
[0100] Step S430: When the real-time operating parameters are abnormal, marking the abnormal circuit of the solid state drive.
[0101] Specifically, when the real-time operating parameters are abnormal, the host computer marks the abnormal circuit of the solid state drive.
[0102] Step S440: Substituting the real-time operating parameters of the abnormal circuit of the solid state drive as input data into the fault prediction model, and the obtained output data is the predicted fault data.
[0103] Specifically, the host computer substitutes the real-time operating parameters of the abnormal circuit of the solid state drive as input data into the fault prediction model, and the obtained output data is the predicted fault data.
[0104] In this embodiment, only the abnormal real-time operating parameters (ie, abnormal parameters) are substituted into the fault prediction model to obtain output data, which can simplify the calculation of the fault prediction model.
[0105] In a fifth embodiment of a method for detecting solid state drive failure based on electrical variables proposed by the present invention, based on the first embodiment, step S140 includes the following steps:
[0106] Step S510: Obtain the importance level and original order of each task in the task queue.
[0107] Specifically, the host computer obtains the importance level and original order of each task in the task queue. The tasks in the task queue include: reading data task, writing data task, modifying data task and deleting data task; the importance level includes two levels: important and ordinary.
[0108] Step S520: reordering the tasks in the task queue of the solid state drive according to the predicted fault type, predicted fault occurrence time, importance level and original order of each task, and marking each task for reordering and original order.
[0109] Specifically, the host computer reorders the tasks in the task queue of the solid state drive according to the predicted fault type, predicted fault occurrence time, importance level of each task and original order, and marks each task for reordering and original ordering.
[0110] For example: the predicted fault type is a power supply circuit fault, and the predicted fault occurrence time is 3 minutes from the current start; then the host computer obtains the estimated completion time of each task in the task queue, and puts the tasks with an estimated completion time of less than 3 minutes (predicted fault occurrence) and an important level at the front of the task queue, and then arranges the tasks with an estimated completion time of less than 3 minutes (predicted fault occurrence) and an ordinary level of importance, and finally arranges other tasks. This can not only prioritize the completion of important tasks, but also ensure that important tasks will not be accidentally interrupted, thereby ensuring the reliability of data in the SSD.
[0111] Step S530: Processing each task in the task queue of the solid state drive according to the order of the reordering mark.
[0112] Specifically, this step is that the solid state drive executes and processes the tasks in the task queue according to the above rearranged order.
[0113] In a sixth embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention, based on the fifth embodiment, step S530 further includes the following steps:
[0114] Step S610: Detect whether the abnormality of the real-time operating parameter is resolved.
[0115] Specifically, the host computer detects whether the abnormality of the real-time operating parameter is resolved.
[0116] If so, execute step S620: process the remaining tasks in the task queue of the solid state drive according to the order of the original sorting marks.
[0117] Specifically, if it is, it means that the abnormal parameters have been resolved, that is, the real-time operating parameters of the solid-state drive are normal, that is, the solid-state drive is operating normally again, then the remaining tasks in the task queue of the solid-state drive can be directly processed in the order of the original sorting marks; to ensure the original processing plan.
[0118] In a seventh embodiment of a method for detecting a fault of a solid state drive based on electrical variables proposed by the present invention, based on any one of the first to sixth embodiments, this embodiment further includes the following steps:
[0119] Step S710: When the host computer in communication with the solid state drive can identify the solid state drive, it is determined whether the host computer can access the cache unit of the solid state drive.
[0120] Specifically, in this embodiment, the solid-state hard disk is a solid-state hard disk that has a fault and needs to be located; the solid-state hard disk is connected to the host computer through the SATA interface. When the host computer can identify the solid-state hard disk, it means that the main control chip of the solid-state hard disk is normal, and then it is further determined whether the host computer can access the cache unit of the solid-state hard disk.
[0121] If so, execute step S720: generate first fault information through the host computer, wherein the first fault information is used to indicate that the fault location of the solid state drive is the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit poor soldering and / or circuit short circuit.
[0122] Specifically, if yes, it means that the cache unit of the solid-state drive is also normal, because the solid-state drive mainly includes three hardware modules, namely the main control chip, the cache unit and the flash memory particles; then when the main control chip and the cache unit of the solid-state drive are normal, and the solid-state drive fails, it means that the fault location is in the flash memory particles, so the first fault information is generated by the host computer, wherein the first fault information is used to express that the fault location of the solid-state drive is the flash memory particle power supply circuit, and the fault causes of the solid-state drive include circuit poor soldering, and / or circuit short circuit.
[0123] If not, execute step S730: generate second fault information through the host computer, wherein the second fault information is used to indicate that the fault location of the solid state drive is the cache unit power supply circuit or the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit poor soldering and / or circuit short circuit.
[0124] Specifically, if no, it means that only the main control chip of the solid-state hard disk is normal, and the fault location may be the cache unit or the flash memory particle; for this purpose, the second fault information is generated by the host computer, wherein the second fault information is used to indicate that the fault location of the solid-state hard disk is the cache unit power supply circuit or the flash memory particle power supply circuit, and the fault causes of the solid-state hard disk include circuit poor soldering and / or circuit short circuit.
[0125] Step S740: When the host computer cannot identify the solid-state hard disk, the host computer generates third fault information, wherein the third fault information is used to indicate that the fault location of the solid-state hard disk is the main control chip power supply circuit, and the fault causes of the solid-state hard disk include circuit poor soldering and / or circuit short circuit.
[0126] Specifically, when the host computer is able to identify the solid-state hard drive, it means that the main control chip of the solid-state hard drive is faulty. For this purpose, the host computer generates third fault information, wherein the third fault information is used to indicate that the fault location of the solid-state hard drive is the main control chip power supply circuit, and the causes of the solid-state hard drive failure include circuit poor soldering and / or circuit short circuit.
[0127] This embodiment provides a technical solution for locating the fault position of a faulty solid state drive.
[0128] In an eighth embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention, based on any one of the first to sixth embodiments, this embodiment further includes the following steps:
[0129] Step S810: obtaining the total operating current of the solid state drive through the detection unit.
[0130] Specifically, in this embodiment, the solid-state hard disk is a solid-state hard disk that has a fault and needs to locate the fault; the solid-state hard disk is connected to the host computer through the SATA interface, and then the host computer obtains the total working current of the solid-state hard disk through the detection unit (preferably a current detection tool in this embodiment). The total working current here is the total current when the solid-state hard disk is working, and the total current is equal to the sum of the working currents of all hardware in the solid-state hard disk.
[0131] Step S820: Sending an erase instruction or a soft destroy instruction to the solid state drive through the host computer, and marking the time when the host computer sends the erase instruction or the soft destroy instruction to the solid state drive as the starting time.
[0132] Step S830: Determine whether the average value of the total operating current in the preset time period after the start time is greater than the average value of the total operating current in the preset time period before the start time.
[0133] Specifically, the host computer determines whether the average value of the total working current in a preset time period (for example, 5 seconds) after the starting moment is greater than the average value of the total working current in the preset time period (for example, 5 seconds) before the starting moment.
[0134] If so, execute step S840: generate fourth fault information by the host computer, wherein the fourth fault information is used to indicate that the fault location of the solid state drive does not include the flash memory particles.
[0135] Specifically, if yes, it means that after the host computer sends an erase command to the SSD, the total current of the SSD increases. That is to say, the flash memory particles in the SSD do execute the erase command or the soft destroy command (the erase command or the soft destroy command are both commands executed by the flash memory particles). This will cause the total current of the SSD to increase. This means that the flash memory particles of the SSD are normal.
[0136] Therefore, the fourth fault information is generated by the host computer, wherein the fourth fault information is used to express that the fault location of the solid state drive does not include the flash memory particles, that is, the fault location of the solid state drive excludes the flash memory particles.
[0137] If not, execute step S850: generate fifth fault information through the host computer, wherein the fifth fault information is used to indicate that the fault location of the solid state drive is the flash memory particle.
[0138] Specifically, if no, it means that after the host computer sends the erase command to the SSD, the total current of the SSD does not increase, that is, the flash memory particles in the SSD do not execute the erase command or the soft destroy command, which means that the flash memory particles in the SSD are faulty.
[0139] Therefore, the fifth fault information is generated by the host computer, wherein the fifth fault information is used to express that the fault location of the solid state drive is the flash memory particle.
[0140] In a ninth embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention, based on the eighth embodiment, step S810 further includes the following steps:
[0141] Step S910: writing test data to the solid state drive through the host computer.
[0142] Specifically, the solid-state hard disk here is a repaired solid-state hard disk, that is, the purpose of the technical solution proposed in this embodiment is to test the repaired solid-state hard disk to determine whether the repair of the solid-state hard disk is complete.
[0143] Step S920: Obtain the writing time required for the test data to be completely written into the solid state drive.
[0144] Specifically, based on the size of the test data and the write speed of the solid state drive, the host computer can obtain the write time (eg, 10 seconds) required for the test data to be completely written into the solid state drive.
[0145] Step S930: Mark all the total operating currents within the write duration as a target current set.
[0146] Specifically, the host computer marks all the total operating currents within the writing time as a target current set; that is, the total operating current here is collected once every 0.1 seconds, so the target current set in this embodiment includes the total operating currents at 100 different times.
[0147] Step S940: obtaining an average value of the total operating current when the solid state drive does not execute any instructions, and marking it as an idle current.
[0148] Specifically, the host computer obtains the average value of the total working current when the solid state hard disk does not execute any instructions, and marks it as the idle current.
[0149] Step S950: Determine whether the proportion of the total working currents whose target current concentration is greater than the idle current is greater than a preset proportion.
[0150] Specifically, the host computer determines whether the proportion of the total working currents in the target current concentration that are greater than the idle current is greater than a preset proportion (for example, 50%).
[0151] If so, execute step S960: generate first detection information, wherein the first detection information is used to indicate that the solid state drive has been repaired normally.
[0152] Specifically, if yes, that is, whether the proportion of the total working currents that are greater than the idle current in the target current concentration is greater than 50%, then when writing test data, the total working current of the solid-state hard disk is greater than the idle current for at least 50 times, which means that during the writing time, the total working current for at least 50% of the time has increased compared to the idle current, then it means that the solid-state hard disk is indeed responding to the data writing function, thereby indicating that the solid-state hard disk has been repaired, so the upper computer generates the first detection information, wherein the first detection information is used to indicate that the solid-state hard disk has been repaired normally.
[0153] In a tenth embodiment of a method for solid state drive fault detection based on electrical variables proposed by the present invention, based on any one of the first to sixth embodiments, this embodiment further includes the following steps:
[0154] Step S1010: remove the flash memory particles of the solid state drive, and establish a communication connection between the removed flash memory particles and the host computer.
[0155] Specifically, in this embodiment, the solid-state hard disk has a fault, but the fault location is not the flash memory particles of the solid-state hard disk, that is, the flash memory particles of the faulty solid-state hard disk can be used normally; after the flash memory particles of the solid-state hard disk are removed separately, a communication connection is established between the removed flash memory particles and the host computer.
[0156] Step S1020: Obtain the remaining erase and write times C and the average daily data write volume R of the flash memory particles of the solid state drive d .
[0157] Specifically, the host computer obtains the remaining erase and write times C and the average daily data write volume R of the flash memory particles of the solid-state drive d . Since the number of times a solid-state drive can be erased and written is limited, the number of times it can be erased determines its service life. The number of erases here refers to the number of times all flash memory particles in the solid-state drive are erased after all data is written into them. For example, if the capacity of the solid-state drive is 500GB and the number of erases is 1000, the actual amount of data that can be erased from this solid-state drive is 500,000GB.
[0158] Step S1030: Calculate the remaining lifespan T of the flash memory particles based on the remaining erase and write times and the average daily data write volume. s :
[0159]
[0160] Wherein, R is the total capacity of the flash memory particles, in GB (e.g., 500 GB), R d The unit is GB (e.g. 50G), T s The unit is day.
[0161] In this embodiment, the remaining erase and write times C is 1, so the remaining life of the flash memory particle calculated is T s For 10 days.
[0162] Step S1040: when the remaining life is less than a preset life, the flash memory particles are scrapped and the data in the flash memory particles are deleted.
[0163] Specifically, when the remaining life is lower than the preset life (for example, 30 days), it indicates that the remaining life of the flash memory particle is relatively short, and it is not worth reinstalling the main control chip and cache unit for repair; therefore, the flash memory particle is directly scrapped and the data in the flash memory particle is deleted to prevent data leakage.
[0164] Step S1050: When the remaining life is not less than the preset life, the flash memory particles are equipped with a normal main control chip and a cache unit.
[0165] Specifically, when the remaining life is not less than the preset life (for example, 30 days), it means that the remaining life of the flash memory particle is still relatively long, and it is worth reinstalling the main control chip and cache unit for repair; therefore, the flash memory particle is directly installed with a normal main control chip and cache unit, and it can be used normally after repair.
[0166] In addition, all flash memory particles have original bad blocks before leaving the factory, which is unavoidable. The appearance of bad blocks will lead to abnormal read and write rates, because the capacity will cause data loss, and the charge jump will cause abnormal read and write data. The main control will reread the flash memory particles and replace the bad blocks. Too many bad blocks will lead to data loss, disk drop, and reduced service life of the solid state drive. If bad sectors appear, it may cause the loss of large-capacity data and storage space of the entire disk.
[0167] However, due to the characteristics of flash memory particles, after multiple reads, writes and erases, the cell barriers of flash memory particles storing charges will become thinner, resulting in read and write data errors or even data loss. In other words, multiple erasing of a particle will lead to the generation of bad blocks and bad sectors, which will in turn reduce the life of the solid-state drive. The main control chip will use erase and write leveling and other means to avoid this situation.
[0168] Flash memory particles are mainly divided into SLC, MLC, TLC, and QLC, and the number of erase and write times of each type is different. SLC can be erased 100,000 times, MLC can be erased 3,000-10,000 times, TLC can be erased 1,000-3,000 times, and QLC can be erased 200-500 times. The above erase and write times are only used as a reference and are not the actual erase and write times of each particle.
[0169] The fact that the entire SSD has a long erase and write life does not necessarily mean that the storage disk performance is good. Flash memory particles are the main carrier of data storage. If there are too many bad blocks or bad sectors in the flash memory particles, it is abnormal and will cause speed loss or data loss. In this case, the flash memory particles need to be replaced.
[0170] To this end, in this embodiment, the number of bad blocks and bad tracks of the flash memory particles can also be detected in real time by the host computer. When the number of bad blocks is greater than the first critical value (e.g., 10%), or the number of bad tracks is greater than the second critical value (e.g., 10%), it indicates that the operation of the flash memory particles is unstable, and it is easy to have read and write speed drops and data loss. It is not worth reinstalling the main control chip and cache unit for repair; therefore, the flash memory particles are directly scrapped and the data in the flash memory particles are deleted to prevent data leakage. The flash memory particles are scrapped and the data in the flash memory particles are deleted.
[0171] The present invention also proposes a system for solid-state hard disk fault detection based on electrical variables. The system applies the method for solid-state hard disk fault detection based on electrical variables as described in any one of the above; the system includes a host computer and a detection unit; the detection unit is communicatively connected to the host computer.
[0172] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0173] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.
[0174] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.
Claims
1. A method for solid state hard disk fault detection based on electrical variables, characterized in that: A system for detecting faults of solid state drives based on electrical variables; the system comprising a host computer and a detection unit; the detection unit is communicatively connected to the host computer; the method comprising: Obtaining a fault prediction model trained by the host computer; Detecting the real-time operating parameters of the solid state drive through the detection unit, and determining whether the real-time operating parameters are abnormal; When the real-time operating parameters are abnormal, the real-time operating parameters of the solid state drive are substituted as input data into the fault prediction model, and the obtained output data is the predicted fault data; reordering the tasks in the task queue of the solid state drive according to the predicted fault type and the predicted fault occurrence time in the predicted fault data; The acquiring of the fault prediction model trained by the host computer includes: Establishing a fault prediction model through the host computer; Perform a fault test to obtain corresponding test fault data after the solid state drive runs for a set time under several groups of test running parameters, wherein the test fault data includes the test fault type and the test fault occurrence time; Using several groups of the test operation parameters as training input data and the corresponding test fault data as training output data to train the fault prediction model; The operating parameters include reset circuit operating parameters, power supply circuit operating parameters and clock circuit operating parameters; the fault types include reset circuit abnormality, power supply circuit abnormality and clock circuit abnormality; the fault test is performed to obtain corresponding test fault data of the solid state drive after running for a set time under several groups of test operating parameters, including: Perform a reset circuit fault test to obtain first test fault data corresponding to the solid state drive after the solid state drive runs for a set time under several sets of reset circuit test operating parameters; Perform a power supply circuit fault test to obtain corresponding second test fault data after the solid state drive runs for a set time under several sets of power supply circuit test operating parameters; A clock circuit fault test is performed to obtain corresponding third test fault data after the solid state hard disk runs for a set time under several sets of clock circuit test operating parameters.
2. The method for solid state hard disk fault detection based on electrical variables according to claim 1, characterized in that: The detecting unit detects the real-time operating parameters of the solid state drive and determines whether the real-time operating parameters are abnormal, including: Respectively detecting the real-time operating parameters of the reset circuit, the power supply circuit and the clock circuit of the solid state drive; respectively judging whether there are abnormalities in the real-time operating parameters of the reset circuit, the real-time operating parameters of the power supply circuit, and the real-time operating parameters of the clock circuit; When the real-time operating parameters are abnormal, the real-time operating parameters of the solid state drive are substituted as input data into the fault prediction model, and the output data obtained is the predicted fault data, including: When the real-time operating parameters are abnormal, marking an abnormal circuit of the solid state drive; The real-time operating parameters of the abnormal circuit of the solid state drive are substituted into the fault prediction model as input data, and the output data obtained is the predicted fault data.
3. The method for solid state hard disk fault detection based on electrical variables according to claim 1, characterized in that: The reordering of the tasks in the task queue of the solid state drive according to the predicted fault type and the predicted fault occurrence time in the predicted fault data includes: Obtaining the importance level and original order of each task in the task queue; Reordering the tasks in the task queue of the solid state drive according to the predicted fault type, predicted fault occurrence time, importance level and original order of each task, and marking each task with a reordering mark and an original order mark; Process each task in the task queue of the solid state drive in the order of the reordering mark.
4. The method for solid state hard disk fault detection based on electrical variables according to claim 3, characterized in that: The tasks in the task queue of the solid state drive are processed according to the order of the reordering marks, and then the steps further include: Detecting whether the abnormality of the real-time operating parameter is resolved; If so, the remaining tasks in the task queue of the solid state drive are processed in the order of the original sorting marks.
5. A method for solid state hard disk fault detection based on electrical variables according to any one of claims 1 to 4, characterized in that: Also includes: When the host computer in communication with the solid state drive can identify the solid state drive, determining whether the host computer can access the cache unit of the solid state drive; If yes, the host computer generates first fault information, wherein the first fault information is used to indicate that the fault location of the solid state drive is the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit poor soldering and / or circuit short circuit; If not, second fault information is generated by the host computer, wherein the second fault information is used to indicate that the fault location of the solid state drive is the cache unit power supply circuit or the flash memory particle power supply circuit, and the fault cause of the solid state drive includes circuit cold soldering and / or circuit short circuit; When the host computer cannot identify the solid-state hard disk, the third fault information is generated by the host computer, wherein the third fault information is used to indicate that the fault location of the solid-state hard disk is the main control chip power supply circuit, and the fault causes of the solid-state hard disk include circuit poor soldering and / or circuit short circuit.
6. A method for solid state hard disk fault detection based on electrical variables according to any one of claims 1 to 4, characterized in that: Also includes: Acquire the total operating current of the solid state hard disk through the detection unit; Sending an erase instruction or a soft destroy instruction to the solid-state hard disk through the host computer that is communicatively connected to the solid-state hard disk, and marking the time when the host computer sends the erase instruction or the soft destroy instruction to the solid-state hard disk as the starting time; Determine whether the average value of the total working current in the preset time period after the starting moment is greater than the average value of the total working current in the preset time period before the starting moment; If yes, fourth fault information is generated by the host computer, wherein the fourth fault information is used to indicate that the fault location of the solid state drive does not include flash memory particles; If not, fifth fault information is generated by the host computer, wherein the fifth fault information is used to indicate that the fault location of the solid state drive is the flash memory particle.
7. A method for solid state hard disk fault detection based on electrical variables according to any one of claims 1 to 4, characterized in that: Also includes: Removing the flash memory particles of the solid state drive, and establishing a communication connection between the removed flash memory particles and the host computer; Get the remaining erase and write times C and the average daily data write volume R of the flash memory particles of the solid-state drive d ; The remaining lifespan T of the flash memory particles is calculated based on the remaining erase and write times and the average daily data write amount. s : Wherein, R is the total capacity of the flash memory particles, in GB; R d The unit is GB, T s The unit of is day; When the remaining life is less than a preset life, the flash memory particles are scrapped and the data in the flash memory particles are deleted; When the remaining life is not less than the preset life, the flash memory particles are equipped with a normal main control chip and a cache unit.
8. A system for detecting solid state hard disk faults based on electrical variables, characterized in that: A method for solid state hard disk fault detection based on electrical variables as described in any one of claims 1 to 7 is applied; the system includes a host computer and a detection unit; the detection unit is communicatively connected to the host computer.
