State monitoring method of solid state disk, electronic device and storage medium

Abstract

The application discloses a state monitoring method of a solid state disk, an electronic device and a storage medium, relates to the technical field of data storage, and comprehensively evaluates the health state of each block storage in the solid state disk through weighted summation of temperature, erasing times, average read-write error rate and other multi-source data, breaks the limitation of isolated analysis of parameters and single analysis dimension in the related monitoring mechanism, and solves the problem that cross-layer data cannot be effectively integrated and analyzed in the related technology, so that the health state of the solid state disk cannot be accurately evaluated and predicted.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data storage, and particularly relates to a state monitoring method of a solid state disk, an electronic device and a storage medium. BACKGROUND

[0002] As a core carrier of modern storage technology, the performance and reliability of a solid state disk (SSD) are highly dependent on the characteristics of NAND (Not AND, for short, AND gate type) flash media and the controller firmware algorithm. The reliability of NAND flash, as the core storage medium of a solid state disk, is significantly affected by charge leakage, read interference and temperature fluctuations.

[0003] Related NAND flash management technology cannot effectively integrate and analyze cross-layer data (such as temperature, P / E cycles, bad block distribution, etc.), making it difficult to accurately assess and predict the health status of an SSD. SUMMARY

[0004] The present application provides a state monitoring method of a solid state disk, an electronic device and a storage medium to at least solve the problem in the related art that cross-layer data cannot be effectively integrated and analyzed, making it difficult to accurately assess and predict the health status of an SSD.

[0005] The present application provides a state monitoring method of a solid state disk, comprising: acquiring multi-source data of NAND gate type flash memory in a to-be-tested solid state disk; the NAND gate type flash memory comprises a plurality of block memories; the multi-source data comprises current erase-write times, average read-write error rates and real-time temperatures of the plurality of block memories; respectively performing weighted summation on the current erase-write times, the average read-write error rates and the real-time temperatures of the block memories in the plurality of block memories to obtain cross-layer state indexes of the block memories in the plurality of block memories; determining a state index of the to-be-tested solid state disk according to the cross-layer state indexes of the block memories in the plurality of block memories, and performing state monitoring on the to-be-tested solid state disk based on the determined state index of the to-be-tested solid state disk.

[0006] The application further provides a state monitoring device of a solid state disk, comprising: a cross-layer collection module, configured to acquire multi-source data of NAND flash in a to-be-tested solid state disk; the NAND flash comprises a plurality of block storages; the multi-source data comprises current erase-write times, average read-write error rates and real-time temperatures of the plurality of block storages; a state prediction module, configured to perform weighted summation on the current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages respectively, to obtain cross-layer state indexes of the block storages in the plurality of block storages; and a state monitoring module, configured to determine a state index of the to-be-tested solid state disk according to the cross-layer state indexes of the block storages in the plurality of block storages, and perform state monitoring on the to-be-tested solid state disk based on the determined state index of the to-be-tested solid state disk.

[0007] The application further provides an electronic device, comprising: a memory configured to store a computer program; and a processor configured to implement the steps of the state monitoring method of the solid state disk when executing the computer program.

[0008] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the state monitoring method of the solid state disk.

[0009] The application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the steps of the state monitoring method of the solid state disk.

[0010] Through the application, multi-source data such as temperature, erase-write times and average read-write error rates are integrated, the health state of each block storage in a solid state disk (SSD) is comprehensively evaluated through weighted summation, the limitation of isolated analysis of each parameter and single analysis dimension in the related monitoring mechanism is broken, and the problem that cross-layer data cannot be effectively integrated and analyzed in the related technology, so that it is difficult to accurately evaluate and predict the health state of the SSD, is solved. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0012] Figure 1 An application environment diagram of the state monitoring method of the solid state disk provided by the embodiments of the application.

[0013] Figure 2 A flowchart of the state monitoring method of the solid state disk provided by the embodiments of the application.

[0014] Figure 3 is a schematic diagram of multiple levels of a to-be-tested solid state disk provided by an embodiment of the present application.

[0015] Figure 4 is a schematic diagram of a voltage threshold curve of a current scanning point provided by an embodiment of the present application.

[0016] Figure 5 is a flowchart of drawing a specified block stored in a voltage threshold curve corresponding to a current scanning point provided by an embodiment of the present application.

[0017] Figure 6 is a structural diagram of a state monitoring device of a solid state disk provided by an embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, any other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0019] It should be noted that, in the description of the present application, the terms “comprise”, “contain” or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device. The terms “first”, “second” and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0020] In order for those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0021] The terms explained in the embodiments of the present application are as follows:

[0022] SSD: Solid State Drive, abbreviated as solid state disk, including a master control chip, a DRAM (Dynamic Random Access Memory, abbreviated as dynamic random access memory) cache and a NAND (Not AND, abbreviated as NAND) particle group of solid state storage device.

[0023] NAND FLASH: a non-volatile storage technology, using “NAND” structure, widely used in solid state disk (SSD), U disk and other devices, its core features are high density, low cost and non-volatility.

[0024] P / E Cycle: Program / Erase Cycle, short for programming / erase cycle, is a core indicator to measure the wear level of a NAND block.

[0025] FTL: Flash Translation Layer, which realizes the mapping of logical address to physical address, contains wear leveling and garbage collection functions.

[0026] REBR: Raw Bit Error Rate, which reflects the physical degradation of the storage unit without error correction.

[0027] SMART: Self-Monitoring, Analysis and Reporting Technology, short for self-monitoring, analysis and reporting technology, is used for SSD health monitoring.

[0028] QLC: Quadruple Level Cell, short for quad-level cell flash memory.

[0029] According to an aspect of the present application, a solid state disk state monitoring method is provided. Optionally, in the present embodiment, the above-mentioned solid state disk state monitoring method can be applied to, but is not limited to, a computer device as shown in Figure 1 , which includes a solid state disk 102 and a processor 104. The processor 104 sends instructions to the solid state disk 102 to request state information such as temperature, programming / erase cycle (P / E Cycle), bit error rate (REBR) and other key indicators. After receiving these requests, the solid state disk 102 provides the corresponding real-time or historical data to the processor 104, which analyzes and displays these data on an interactive visualization platform, so that users or systems can monitor the state information of the solid state disk and take timely maintenance measures or optimization strategies.

[0030] The above-mentioned computer device can be, but is not limited to, a personal computer (PC), a mobile phone, a tablet computer, a cloud server, a server cluster or other server types.

[0031] The solid state disk state monitoring method of the present application can be executed by a computer device, Figure 2 is a flowchart of an optional solid state disk state monitoring method according to an embodiment of the present application, as shown in Figure 2 , the flow of the method can include the following steps:

[0032] In step S202, multi-source data of NAND flash memory in the to-be-tested solid state disk is acquired; the NAND flash memory includes a plurality of block storages; the multi-source data includes current erase-write times, average read-write error rates and real-time temperatures of the plurality of block storages.

[0033] In step S204, the current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages are respectively weighted and summed to obtain cross-layer state indexes of the block storages in the plurality of block storages.

[0034] In step S206, a state index of the to-be-tested solid state disk is determined according to the cross-layer state indexes of the block storages in the plurality of block storages, and state monitoring is performed on the to-be-tested solid state disk based on the determined state index of the to-be-tested solid state disk.

[0035] Embodiments of the present application provide a solid state disk state monitoring method, which is used in the field of data storage technology, and specific scenarios include but are not limited to health management of solid state disks in enterprise servers, data centers, cloud computing storage systems and high-performance computing environments. By analyzing multi-source data of SSD in real time, potential failures are predicted, storage allocation is optimized, data loss risk is reduced, and the reliability and efficiency of the overall storage system are improved.

[0036] As the core carrier of modern storage technology, the performance and reliability of a solid state disk (SSD) are highly dependent on the characteristics of NAND flash memory media and the controller firmware algorithm. NAND flash memory, as the core storage medium of a solid state disk, its reliability is significantly affected by charge leakage, read interference and temperature fluctuations. Related NAND flash memory management technologies cannot effectively integrate and analyze cross-layer data (such as temperature, P / E cycles, bad block distribution, etc.), making it difficult to accurately assess and predict the health status of the SSD.

[0037] Therefore, to solve the above problems, in the embodiments of the present application, multi-source data is collected across layers, and the state of the SSD is monitored based on the multi-source data. The multi-source data refers to comprehensive information about the health status of NAND flash memory collected from a plurality of levels of the to-be-tested solid state disk. The current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages can be weighted and summed by a pre-trained health index model to obtain cross-layer state indexes of the block storages in the plurality of block storages. Wherein, Figure 3 A schematic diagram of a plurality of levels of a to-be-tested solid state disk provided in the embodiments of the present application is shown in Figure 3As shown, the multiple levels of the to-be-tested solid state disk include SMART data, physical layer signals, and FTL metadata, each data type reflecting different aspects of the state of the flash memory, collectively forming a detailed health monitoring framework. The SMART data is part of the solid state disk self-monitoring, analysis, and reporting technology, providing basic health indicators of the solid state disk, including real-time temperature, read / write erase failure count, and percentage of remaining life. The physical layer signals relate to the underlying physical characteristics of the NAND flash memory, particularly the voltage threshold (Vth) distribution obtained through offset voltage scanning. The Vth distribution can reveal physical degradation of the memory cells, such as charge leakage, read interference, etc., and is an important window for in-depth understanding of the internal health of the NAND flash memory. By combining with the FTL metadata, the health index (HI) of each Block can be calculated, which embodies the direct contribution of the physical layer data to the health assessment of the solid state disk. The FTL metadata belongs to the flash translation layer, which plays a bridge role in the solid state disk, storing details about bad blocks, program / erase cycles (P / E Cycle), error code rate (REBR), etc., which are key indicators reflecting the wear and degradation of the NAND flash memory. The FTL metadata not only contains bad block information at the factory, but also includes newly added bad block records during operation, as well as detailed P / E and REBR statistics at the Block level.

[0038] In related technologies, the SMART log and debug interface data are not prioritized, and the data collection is extensive, resulting in serious resource waste. Therefore, to solve this problem, in the embodiments of the present application, differentiated collection frequencies are proposed for different data types, such as collecting data at different frequencies according to their importance, unifying heterogeneous data, aligning them according to timestamps, and saving them to a local database, which ensures the real-time nature of key monitoring indicators and avoids resource waste, solving the problems of extensive and low-efficiency data collection in related technologies. The multi-source data of the present application includes but is not limited to the temperature, program / erase cycle (P / E Cycle), bad block distribution, raw error code rate (REBR) stored in each block, etc. The collection strategy of the multi-source data is shown in Table 1 as follows:

[0039] Table 1

[0040]

[0041] Among them, the current erase-write cycle (P / E Cycle) refers to the total number of programming (writing) and erasing operations that a certain block (Block) in the NAND Flash has undergone since production. The average read-write error rate (Raw Bit Error Rate, REBR) is an indicator that measures the error proportion between the original data and the actual stored data of the NAND Flash storage unit during read or write operations. In this application, it is used to evaluate the physical degradation of the storage block (Block), and the higher the average REBR, the lower the reliability of the storage unit. Real-time temperature refers to the instantaneous temperature reading of the internal NAND gate type flash memory of the solid state disk during operation. Temperature has a significant impact on the performance and lifespan of NAND Flash, and both excessively high and low temperatures will accelerate the degradation of the storage unit and reduce data reliability.

[0042] NAND gate type flash memory specifically refers to a non-volatile storage technology that uses "NAND" circuit architecture, with characteristics of high-density storage, fast read-write, and low power consumption, widely used in solid state disks (SSD). The solid state disk to be tested contains NAND gate type flash memory, which is composed of multiple block storage units. Block storage is the basic unit of data storage in NAND Flash, each block is composed of multiple pages (Page) and can be independently erased. The performance and reliability of the solid state disk are highly dependent on the state of the block storage unit, especially the current erase-write cycle, average read-write error rate, and real-time temperature parameters. This application evaluates the health status of the entire solid state disk by monitoring these multi-source data to achieve fault prediction and management.

[0043] The cross-layer state index is a comprehensive health indicator that evaluates the current state and potential risks of NAND Flash block storage. It intelligently weights and sums key parameters such as the current erase-write cycle, average read-write error rate, and real-time temperature of the block storage to form a quantitative value, directly reflecting the wear and degradation of the storage block in different dimensions. The weight is dynamically adjusted according to the impact of each parameter on storage reliability, making the evaluation result more close to the actual situation, effectively predicting and diagnosing early failures of the storage unit, and thus optimizing the management of the SSD and prolonging its service life.

[0044] The state index is a quantitative indicator of the overall health status of the solid state disk (SSD) to be tested. It is based on the comprehensive analysis of the cross-layer state index, i.e. starting from the block storage (Block) level of the NAND Flash, considering key parameters such as temperature, current erase-write cycle (P / E Cycle), and average read-write error rate (REBR), and obtaining the health status score (HI) of each block by weighted summation. The state index further integrates these block-level HI values to reflect the wear, degradation, and operating environment health status of the entire SSD.

[0045] By the embodiments of the present application, multi-source data such as temperature, erasing times, average read-write error rate, etc. are integrated, and the health status of each block storage in a solid state disk (SSD) is comprehensively evaluated through weighted summation, breaking the limitation of isolated analysis of each parameter and single analysis dimension in related monitoring mechanisms, and solving the problem that cross-layer data cannot be effectively integrated and analyzed in related technologies, so that it is difficult to accurately evaluate and predict the health status of the SSD.

[0046] In one example embodiment, the current erasing times, the average read-write error rate and the real-time temperature of the block storage in the plurality of block storages are respectively weighted and summed to obtain a cross-layer state index of the block storage in the plurality of block storages, comprising:

[0047] I. The ratio between the current erasing times of the block storage in the plurality of block storages and the maximum tolerable erasing times of the block storage in the plurality of block storages is calculated to obtain a first ratio corresponding to the block storage in the plurality of block storages.

[0048] II. The ratio between the average read-write error rate of the block storage in the plurality of block storages and the maximum read-write error rate of the block storage in the plurality of block storages is calculated to obtain a second ratio corresponding to the block storage in the plurality of block storages.

[0049] III. The temperature difference between the real-time temperature of the block storage in the plurality of block storages and the optimal working temperature of the block storage in the plurality of block storages is calculated, and the ratio between the temperature difference of the block storage in the plurality of block storages and the maximum allowed temperature of the block storage in the plurality of block storages is calculated to obtain a third ratio corresponding to the block storage in the plurality of block storages.

[0050] IV. The first ratio corresponding to the block storage in the plurality of block storages, the second ratio corresponding to the block storage in the plurality of block storages and the third ratio corresponding to the block storage in the plurality of block storages are weighted and summed to obtain a cross-layer state index of the block storage in the plurality of block storages.

[0051] In the present application, the maximum tolerable erasing times refer to the maximum number of programming / erasing (P / E) operations that each block storage unit (Block) in the NAND Flash can withstand in its life cycle, and beyond this number, the reliability of the unit will decrease significantly, which may cause data errors or block failure.

[0052] The first ratio refers to the ratio of the current erasing times of the block storage to the maximum tolerable erasing times, which is used to quantify the wear degree of the block storage. When calculating the cross-layer state index, the higher the first ratio, the closer the block storage is to its life limit, and the greater the negative impact on the health index.

[0053] Maximum read / write error rate is the highest acceptable level of data read / write error in NAND Flash, which reflects the highest error rate that may occur during data read / write operations without taking additional error correction measures. This parameter is compared with the average read / write error rate of the block storage to calculate the second ratio, reflecting the degree of degradation during data read / write process.

[0054] The second ratio is the ratio between the average read / write error rate of the block storage and the maximum read / write error rate, used to evaluate the data integrity status of the block storage. In the calculation of the cross-layer state index, the second ratio reflects the impact of error codes on the health of the block storage, and a ratio close to 1 indicates that the error rate of the block storage is close to its danger threshold, with a questionable health status.

[0055] The optimal operating temperature refers to the best working environment temperature considered in the design of NAND Flash block storage. Within the optimal operating temperature range, the storage device can exhibit the best performance and longest service life, while maintaining the lowest power consumption and most stable reliability. The selection of the optimal operating temperature takes into account factors such as material properties, manufacturing processes, and workload, aiming to provide an ideal operating environment for solid-state drives, reducing performance degradation and data errors caused by temperature. For example, the optimal operating temperature can be 45 degrees.

[0056] The maximum allowed temperature is the highest temperature limit at which NAND Flash can operate stably without causing performance degradation or reliability damage. For example, the maximum allowed temperature can be 60 degrees. Exceeding the maximum allowed temperature, the charge retention capability of the storage medium will be affected, which in turn affects the stability and durability of the data. In the calculation of the third ratio, the ratio of the real-time temperature to the maximum allowed temperature is used to evaluate the temperature risk of the storage environment.

[0057] The third ratio is the ratio of the temperature difference between the real-time temperature and the optimal working temperature to the highest allowable temperature, which reflects the degree of temperature deviation from the optimal state. In this embodiment, the third wallpaper is regarded as the temperature deviation ratio, which represents the degree of deviation of the working temperature of the solid state disk to be tested from the optimal working temperature, and is a standardized ratio relative to the highest allowable temperature. The third ratio will significantly increase in a high temperature environment, triggering the increase of the temperature weight, so as to pay more attention to the temperature factor in the calculation of the cross-layer state index, and to give an early warning of potential temperature risks. The temperature difference refers to the difference between the real-time temperature of the block storage and the optimal working temperature, which is used to quantify the deviation of the current working environment from the ideal state. In this embodiment, the temperature difference between the real-time temperature and the optimal working temperature is used, and then the ratio is calculated with the highest allowable temperature to obtain the third ratio. This method can more accurately measure the degree of deviation of the current temperature of the block storage from the optimal working temperature through the temperature difference, and can more subtly reflect the potential impact of temperature fluctuations on the health status of the NAND Flash compared with directly using the real-time temperature. Further, the ratio of the temperature difference to the highest allowable temperature is used as the third ratio. When the temperature is closer to the highest allowable temperature, the ratio of the temperature difference to the highest allowable temperature is larger, which means that the block storage is facing higher risks. This method can automatically amplify the weight of temperature when the temperature is close to the limit value.

[0058] In this embodiment, the ratio between the current parameter value and the corresponding maximum value is calculated, and then weighted summation is performed, instead of directly performing weighted summation on the current erase-write times, average read-write error rate and real-time temperature, for the following reasons: 1) Standardized data: By calculating the ratio, parameters of different scales and units (such as erase-write times, error rate and temperature) can be converted to a common evaluation scale, usually between 0 and 1. This can avoid the weight distortion caused by the large difference in parameter magnitude when directly adding, and ensure the fairness and comparability of each parameter in the evaluation. 2) Clear physical meaning: The ratio calculation can better reflect the deviation of the parameter from its ideal state (maximum tolerance erase-write times, maximum read-write error rate, highest allowable temperature), thereby directly reflecting the health status of the three key dimensions of the block storage, i.e. wear, data integrity and environmental suitability. This method makes the interpretation of the health index easier to understand, and can be directly related to the life, reliability and environmental adaptability of the block storage. 3) Avoid the influence of data magnitude: The erase-write times may reach several thousand or even several ten thousand, while the values of temperature and error rate are relatively small. If not standardized, direct weighted summation may make the erase-write times dominate in the health index, ignoring the influence of temperature and error rate. The ratio processing ensures that each parameter contributes to the health index according to its deviation from the maximum value, avoiding the problem of single parameter dominance.

[0059] By introducing the first ratio (i.e. the ratio between the current erase-write times of the block storage and the maximum tolerance erase-write times), the second ratio (i.e. the ratio between the average read-write error rate of the block storage and the maximum read-write error rate) and the third ratio (i.e. the ratio between the temperature difference between the real-time temperature and the optimal working temperature and the maximum allowed temperature), the first ratio breaks the limitation of relying only on the current erase-write times for evaluation in the related monitoring mechanism, and the proportion of the erase-write times relative to the life limit is quantified, which more accurately reflects the wear state of the block storage; by calculating the second ratio, the influence of the error code in the data read-write process on the storage health can be more accurately captured, the problem of low efficiency in the related art is solved, a means for quickly identifying the root cause of the error code problem is provided, and long-time fault positioning is avoided; the third ratio emphasizes the importance of temperature on the health of the NAND flash memory, and solves the problem of ignoring the temperature sensitivity of the data acquisition strategy in the related art; the first ratio, the second ratio and the third ratio are summed by weighting to obtain the cross-layer state index, which avoids the problem of inconsistent dimension and order of magnitude caused by directly using the original data, and ensures the equal role of different source data in evaluation.

[0060] In one example embodiment, the above-mentioned state monitoring method of the solid state disk further comprises:

[0061] The block storage in the plurality of block storages is respectively taken as a current block storage to perform the following weight adjustment operation: in the case that the real-time temperature of the current block storage is greater than a preset temperature threshold, the initial weight of the third ratio corresponding to the current block storage is adjusted to obtain a first weight; the first weight is positively correlated with the real-time temperature of the current block storage and is less than a preset weight threshold; according to the first weight, the initial weight of the second ratio corresponding to the current block storage is adjusted to obtain a second weight, and according to the first weight, the initial weight of the first ratio corresponding to the current block storage is adjusted to obtain a third weight; the sum of the first weight, the second weight and the third weight is equal to 1; wherein the weighted sum of the first ratio corresponding to the block storage in the plurality of block storages, the second ratio corresponding to the block storage in the plurality of block storages and the third ratio corresponding to the block storage in the plurality of block storages is performed according to the first weight, the second weight and the third weight.

[0062] In the embodiment, the cross-layer state index of the block storage can be represented by the following formula (1):

[0063]

[0064] wherein, is the maximum tolerance erase-write times of the specified block storage (Block); is the current erase-write times of the specified block storage (Block); is the average read-write error rate of the page that has been read in the specified block storage (Block); to specify the maximum read / write error rate for a specified block storage (Block); to specify the real-time temperature for a specified block storage (Block); to specify the optimal operating temperature for a specified block storage (Block), for example, 45 degrees; to specify the maximum allowed temperature for a specified block storage (Block), for example, 60 degrees; to represent the third weight; to represent the second weight; to represent the first weight, and + + = 1.

[0065] The current block storage refers to a specific NAND Flash memory block that is undergoing health status evaluation. The preset temperature threshold refers to the warning limit value set for the real-time temperature of the block storage in the health monitoring system of the solid-state disk. When the real-time temperature of the block storage exceeds this threshold, the system adjusts the initial weight of the third ratio, indicating that the storage device is in an unfavorable temperature environment and may face the risk of performance degradation or shortened lifespan. This threshold is determined according to the specific temperature sensitivity and optimal working conditions of the NAND Flash material. For example, the preset temperature threshold can be 65 degrees.

[0066] The initial weight of the third ratio is the initial weight value reflecting the impact of the temperature dimension on the health status of the block storage. The initial weight of the third ratio is preset through experiments or experience, representing the default importance measure of the REBR parameter when the temperature does not exceed the maximum allowed temperature. In the absence of special temperature conditions, the initial weight of the third ratio reflects the basic weight of temperature in the calculation of the cross-layer state index. The first weight is the result of adjusting the initial weight of the third ratio when the real-time temperature changes. The first weight is positively correlated with the real-time temperature, but is limited by the preset weight threshold to avoid the excessive influence of the temperature single parameter on the overall evaluation in the health index. The preset weight threshold is an upper limit for the adjustment of the first weight, preventing the weight from being too high due to significant changes in temperature, thereby affecting the overall balance of the health index. For example, the preset weight threshold can be 0.4. The preset weight threshold ensures that regardless of the changes in the real-time temperature, the contribution of the temperature parameter to the health index will not exceed the pre-set limit value.

[0067] Optionally, the computer device continuously collects real-time temperature data of the current block storage through the SSD physical layer interface at a frequency of 2 Hz, sets a preset temperature threshold (such as 65 degrees), and triggers the weight adjustment mechanism when the real-time temperature exceeds the preset temperature threshold. For example, a linear growth model is used, and the first weight is increased correspondingly when the temperature increases by a certain number of degrees, but it is ensured that the first weight is less than a preset weight threshold (such as 0.4). Based on the first weight value, the initial weights of the second ratio (such as the PE period related weight) and the third ratio (such as the REBR related weight) are adjusted in proportion. For example, if the first weight increases, the second and third weights are appropriately reduced to maintain the total weight sum of 1 and ensure the rationality of the weight distribution.

[0068] Through the present embodiment, in the case where the real-time temperature of the current block storage is greater than the preset temperature threshold, the initial weight of the third ratio corresponding to the current block storage is adjusted to obtain the first weight, the first weight is positively correlated with the real-time temperature of the current block storage and is less than the preset weight threshold, the second weight and the third weight are dynamically adjusted according to the change of the first weight, and the total weight sum is ensured to be 1, which can accurately reflect the physical degradation impact of high temperature on the block storage.

[0069] In one exemplary embodiment, in the case where the real-time temperature of the current block storage is greater than the preset temperature threshold, the initial weight of the third ratio corresponding to the current block storage is adjusted to obtain the first weight, including:

[0070] In the case where the real-time temperature of the current block storage is greater than the preset temperature threshold, the temperature difference between the real-time temperature of the current block storage and the maximum allowed temperature is determined as the real-time temperature difference, and the ratio between the real-time temperature difference and the preset weight influence coefficient is determined as the temperature change ratio. According to the temperature change ratio, the first weight adjustment factor is determined, and the product between the first weight adjustment factor and the initial weight of the third ratio of the current block storage is determined as the first weight.

[0071] Wherein, the real-time temperature difference refers to the difference between the real-time temperature of the current block storage and the maximum allowed temperature, which is used to quantify the degree of temperature deviation from the safety range. For example, when the threshold is 65 degrees and the real-time temperature is 85 degrees, the real-time temperature difference is 20 degrees, which directly drives the weight adjustment calculation to ensure the risk response sensitivity in the high temperature scenario.

[0072] The temperature change ratio is determined by the ratio of the real-time temperature difference to the preset weight influence coefficient, which reflects the quantitative impact of temperature change on weight adjustment. The preset weight influence coefficient refers to a fixed parameter for quantifying the impact of the real-time temperature difference on the weight adjustment, and its value is set through experiment or experience as the denominator of the temperature change ratio calculation. For example, if the preset weight influence coefficient is 30 and the real-time temperature difference is 20 degrees, the temperature change ratio is 0.67. The temperature change ratio determines the amplitude of the first weight adjustment factor, realizing the nonlinear correlation between temperature and weight.

[0073] The first weight adjustment factor is a dynamic coefficient calculated according to the temperature change ratio, used to correct the initial weight. For example, the first weight adjustment factor can be a dynamic adjustment coefficient obtained by adding a constant 1 to the temperature change ratio. In this embodiment, the first weight can be expressed by the following formula (2):

[0074]

[0075] wherein the maximum allowable temperature can be 60 degrees, and the preset weight influence coefficient can be 30, represents the initial weight of the third ratio of the current block storage, represents the adjusted weight of the initial weight of the third ratio of the current block storage, i.e. the first weight.

[0076] Through this embodiment, in the case that the real-time temperature of the current block storage is greater than the preset temperature threshold, the temperature difference between the real-time temperature of the current block storage and the maximum allowable temperature is determined as the real-time temperature difference, which directly reflects the physical degradation pressure of the block storage by quantifying the deviation degree of the current temperature from the maximum allowable temperature; the ratio between the real-time temperature difference and the preset weight influence coefficient is determined as the temperature change ratio, which normalizes the temperature difference and the preset weight influence coefficient through the temperature change ratio, eliminates the influence of hardware model differences on the calculation, and ensures the comparability of weight adjustment of different SSDs.

[0077] In one example embodiment, according to the first weight, the initial weight of the second ratio corresponding to the current block storage is adjusted to obtain the second weight, comprising:

[0078] According to the first weight, the current remaining ratio of the first weight relative to the specified weight state is determined, and according to the initial weight of the third ratio corresponding to the current block storage, the original remaining ratio of the initial weight of the third ratio corresponding to the current block storage relative to the specified weight state is determined; the ratio between the current remaining ratio and the original remaining ratio is determined as the second weight adjustment factor, and the product between the second weight adjustment factor and the initial weight of the second ratio corresponding to the current block storage is determined as the second weight.

[0079] wherein the specified weight state refers to the case that the sum of all parameter weights is 1, serving as the reference for weight adjustment. The specified weight state represents the original importance distribution of each parameter in the health index model under normal or ideal operating conditions, and is used as a reference point for subsequent weight changes.

[0080] The current remaining ratio refers to the difference between the first weight and 1 (i.e. 1- ), which reflects the proportion of the remaining space of other parameter weights after the temperature weight adjustment, is used to calculate the second weight adjustment factor to realize the dynamic balance of the weight system.

[0081] The original residual proportion is the difference between the initial weight of the third ratio and 1 when the current block storage is not affected by temperature (i.e., 1- ), which shows the proportion of the relative residual contribution of the temperature parameter to the health index calculation under normal conditions, for comparison with the effect of weight adjustment after temperature change.

[0082] The second weight adjustment factor is calculated based on the ratio of the current residual proportion and the original residual proportion, and is used to dynamically adjust the initial weight of the second ratio (such as the error rate weight) corresponding to the current block storage. The second weight adjustment factor ensures that the importance of key parameters such as read-write error rate can be automatically adapted under temperature-sensitive conditions, maintaining the accuracy and stability of the health index model.

[0083] In this embodiment, the adjusted second weight can be represented by the following formula (3):

[0084]

[0085] In an example embodiment, the adjustment method of the third weight is consistent with that of the second weight, which will not be repeated here.

[0086] In an example, it is assumed that the optimal working temperature of the current block storage is 45 degrees, the maximum allowed temperature is 60 degrees, the maximum endurance of the current block storage =1000, the current endurance of the current block storage =400, the maximum read-write error rate in the current block storage =1.1x10-3, the third weight =0.5, the second weight , and the first weight .

[0087] If the real-time temperature is 45 degrees under normal temperature conditions, which does not exceed the maximum allowed temperature, the first weight remains unchanged, and the average read-write error rate of the page that has been read in the current block storage =7x10-5, then the cross-layer state index of the current block storage can be represented by the following formula (4):

[0088]

[0089] The above formula (4) indicates that the health status of the current block storage is good (HI=0.221).

[0090] If the real-time temperature is 65 degrees, does not exceed the maximum allowable temperature, the first weight remains unchanged, and the average read-write error rate of the pages read in the current block storage = 4x10-4, then the cross-layer state index of the current block storage can be represented by the following formula (5):

[0091]

[0092] The above formula (5) indicates that the current block storage is at medium risk (HI = 0.387).

[0093] If the real-time temperature is 85 degrees, exceeds the maximum allowable temperature, the first weight, the second weight, and the third weight all need to be adjusted, assuming that the preset weight threshold is 0.4, the first weight can be adjusted by the following formulas (6) and (7), the second weight can be adjusted by the following formula (8), and the third weight can be adjusted by the following formula (9):

[0094]

[0095]

[0096]

[0097]

[0098] Assuming that the average read-write error rate of the pages read in the current block storage = 9x10-4, then the cross-layer state index of the current block storage corresponding to the weight adjustment can be calculated by the following formula (10):

[0099]

[0100] The above formula (10) indicates that the current block storage is at high risk (HI = 0.616).

[0101] Through the embodiment, the current remaining proportion of the first weight relative to the specified weight state is calculated, the weight of the temperature parameter in the health index calculation is dynamically adjusted, when the temperature exceeds the preset temperature threshold, the temperature influence proportion is automatically increased, the rapid response to the temperature sensitivity change is realized, the initial weight of the third ratio relative to the original remaining proportion of the specified weight state can quantify the difference before and after the temperature weight adjustment, the second weight is determined by multiplying the second weight adjustment factor and the initial weight of the second ratio, the weight of important indicators such as read-write error rate can be adaptively adjusted according to the real-time temperature change, when the first weight increases due to the temperature rise, the weight of other parameters is automatically reduced, the stability and accuracy of the overall evaluation system are maintained.

[0102] In one example embodiment, the fault location in the related art takes a long time of up to several weeks, therefore, to solve the technical problem, in the embodiment, an interactive visualization platform is provided, which visually displays the physical structure of the SSD, supports time axis backtracking and data drilling, supports multi-chart linkage analysis, thereby improving the fault location speed. The above-mentioned state monitoring method of the solid state disk further comprises:

[0103] According to the physical structure of the to-be-tested solid state disk, a multi-level addressing heat map is constructed; a group of target block storages is located and displayed through the multi-level addressing heat map; the group of target block storages refers to at least one block storage with a cross-layer state index greater than a preset health index; wherein the multi-level addressing heat map is a multi-level view constructed by taking the cross-layer state index of the block storage as a heat value; the multi-level addressing heat map includes a channel level view, a target level view, a logical unit level view and a block level view; the channel level view displays a health status summary of the channel level in the to-be-tested solid state disk; the target level view displays a health status summary of the target level in the to-be-tested solid state disk; the logical unit level view displays the cross-layer state index of the block storage under the logical unit level in the to-be-tested solid state disk; the block level view displays the attributes of the word line under the block level in the to-be-tested solid state disk; the word line is a control line for selecting a specific storage unit.

[0104] Wherein, the multi-level addressing heat map is a visualization tool based on the physical structure of the solid state disk (SSD), which takes the cross-layer state index (HI) as the heat value, and displays the health status of each part of the SSD through the views of different levels (channel, target, logical unit, block), helping to quickly locate the problem area. Inside the solid state disk (SSD), NAND flash memory is organized in a complex and hierarchical structure, specifically, the internal physical structure of the SSD is composed of Channel -> Target -> LUN -> Plane -> Block levels. In the physical architecture of the SSD, Channel is the outermost organizational structure, each Channel contains multiple Targets, each Target manages several LUNs, and each LUN is divided into multiple Planes, and finally, each Plane is composed of multiple Blocks.

[0105] The channel level view is part of the multi-level addressing heat map, which is used to show the health status summary of the channel level in the SSD. The health status summary of the channel level provides a comprehensive and detailed health status overview of the SSD at the Channel level. For example, the health status summary of the channel level includes the HI value (the average of all acquired HI values under this Channel), the number of problem Targets, LUNs, Blocks, and bad blocks. Assuming that the SSD under test includes 16 Channels, the channel level view is presented in the form of a 16x1 matrix, with each cell reflecting the health status summary of the HI value average of a Channel, facilitating a macro view of the overall health status of the Channels. Figure 1

[0106] The target level view is also part of the multi-level addressing heat map, which focuses on the health status summary of the Target level in the SSD. The health status summary of the Target level provides a comprehensive and detailed health status overview of the SSD at the Target level. For example, the health status summary of the Target level includes the HI value (the average of all acquired HI values under this Target), the number of problem Targets, LUNs, Blocks, and bad blocks. Assuming that the SSD under test includes 4 Targets per Channel, the health indicators of the 4 Targets are presented in the form of a 4x1 matrix, providing data support for middle-level analysis.

[0107] The logical unit level view is responsible for displaying the cross-layer state index (HI) of all block storage under each logical unit (LUN) in the multi-level addressing heat map system. All Blocks in this LUN are distributed according to the Plane, and are converted into a two-dimensional grid according to their physical topology. The two-dimensional grid form is adopted, with the X-axis representing the Plane number (e.g. 0~3) and the Y-axis representing the Block number (e.g. 0~M). Each grid cell represents a single Block (the size is dynamically adjusted according to the scaling level), and the color depth of the grid cell reflects the HI value. Bad blocks are marked with different colors according to the out-of-box bad blocks and newly added bad blocks, facilitating detailed health status analysis at the LUN level. In addition to displaying the cross-layer state index of the block storage at the logical unit level, the logical unit level view also displays the attributes of each Block, including the physical address, LBA address, HI, PE, REBR (average), etc. When focusing on a certain Block in the logical unit level view, all the aforementioned attribute values are displayed.

[0108] ​The block level view is the lowest level view of the heat map, which displays the health attributes of a single block through a matrix of wordline number x 1, including wordline level physical address, REBR information, etc. It supports wordline level Vth distribution scanning, which is used for accurate positioning and analysis of block internal health problems. Wordline is a control line used to select specific memory cells or pages on the NAND flash chip in the SSD for read and write operations. It is the basis for building the block level view. By monitoring the attributes of the word line, we can analyze and locate the small changes in the memory cells, thereby ensuring the stability and reliability of the SSD.

[0109] In this embodiment, a set of target block storage specifically refers to a block storage set whose cross-layer health index (HI) exceeds a preset health threshold, and at least contains one block storage. In this embodiment, HI values and REBRs in different statistical ways can be displayed at different levels, and a high-risk block distribution heat map can be displayed at the lun level, and a list of the top 10 (TOP10) deteriorated blocks with the highest HI values can be output.

[0110] This embodiment can bind real-time data and historical data to each view level of the visualization system and associate relevant information data. For example, when a specific wordline is located, a Vth scan is initiated, a curve is drawn, historical scan records are provided, and historical curves can be superimposed. Different types of alarm conditions can also be configured according to usage requirements, and when triggered, relevant data information is automatically recorded and corresponding operations (such as Vth scanning) are performed. The corresponding problem can be popped up or marked in the visualization system.

[0111] In one example, when a certain QLC SSD is under high-temperature stress testing, the method of this embodiment is used to configure an alarm when all Block HI values are greater than 0.7. When the temperature reaches 85°C during testing, the heat map shows that the HI value of CH2-T1-L0 rises to 0.82, which is 14 days earlier than the SMART warning to discover the charge leakage problem. The visualization system accurately locates the specific position, analyzes the Vth distribution of a large number of wordlines in this block, and finds that all the offsets are large. By analyzing the REBR curve associated with reading, it is found that the REBR of this part is high, and the problem is alleviated after enhancing the heat dissipation contrast.

[0112] Through this embodiment, the multi-level addressing heat map is organized in the hierarchy of Channel, Target, LUN, Block and Wordline, so that the fault diagnosis can start from the macro Channel level and gradually refine to the specific Wordline level. This top-down analysis path greatly accelerates the problem positioning process. The heat map visually represents the health status (HI value) of different levels by color depth or temperature, so that the maintenance personnel can immediately identify the fault point without going deep into the code or log file.

[0113] In an example embodiment, as the storage density improves (such as QLC technology), the cell charge capacity decreases, and a slight shift in the voltage threshold distribution can cause a read error. Therefore, in order to monitor the state of the solid state disk under test, the state monitoring method of the above-mentioned solid state disk further includes:

[0114] Initiating a scan instruction package for the specified block storage within the block level view; the scan instruction package includes voltage threshold scan configuration parameters; the scan instruction package is used to instruct the solid state disk under test to scan the specified block storage according to the voltage threshold scan configuration at the current scan point and return the scan data corresponding to the current scan point; according to the scan data, the voltage threshold curve corresponding to the current scan point of the specified block storage is drawn; the voltage threshold curve corresponding to the current scan point is used to analyze the read state and data reliability of the NAND flash corresponding to the current scan point.

[0115] Among them, this embodiment supports nand wordline level Vth scanning and generates Vth distribution curve diagram, supports superposition comparison, difference calculation, etc.

[0116] The specified block storage refers to one or more Block (blocks) in the physical storage structure of the solid state disk (SSD) that are specially selected for monitoring, analysis or operation. The specified block storage can be a block storage at any position in the SSD. For example, the specified block storage can be a target block storage.

[0117] The scan instruction package is a series of instructions and configuration parameters sent by the visualization system to the SSD host, which aims to perform Vth (voltage threshold) distribution scanning on the specified NAND Flash block storage. For example, the scan instruction package contains key information such as target NAND physical address, target Vth value (generally all Vth are scanned, QLC has 15), offset (offset) step (such as 4), and scan range (such as -64~+64), to ensure that the SSD host can accurately perform Vth scanning and return the required data.

[0118] The current scanning point usually refers to the moment when the system issues the Vt distribution scanning instruction and performs the read operation. The voltage threshold curve, i.e., the Vth curve, is a chart generated according to the Vth distribution scanning data, which is used to intuitively show the voltage threshold distribution of each memory cell in the NAND Flash. The curve is crucial for analyzing the physical degradation of the cells (such as charge leakage) and evaluating the reliability of the read operation. Through analysis of the curve, the health status of the NAND Flash block can be discovered and evaluated in a timely manner. Figure 4 is a schematic diagram of the voltage threshold curve of the current scanning point provided by the present embodiment, as shown in Figure 4 The voltage threshold curve of the current scanning point includes a scanning range of -64~+64.

[0119] Optionally, Figure 5 is a flowchart of drawing the voltage threshold curve of the current scanning point in which the specified block is stored, provided by the present embodiment, as shown in Figure 5 The visualization system in the computer device sends a scanning instruction package to the SSD host chip, wherein the scanning instruction package includes a target Vth value (such as RL10_QLC, which refers to the 10th read level for QLC (four-layer cell flash)), an offset step (such as 4), and a scanning range (such as -64~+64). The SSD host cyclically scans the NAND chip according to the set offset step, a total of 65 times, and the NAND chip returns the scanning data to the SSD host. The SSD host forwards the scanning data to the visualization system. The visualization system draws the voltage threshold curve of the current scanning point in which the specified block is stored according to the scanning data.

[0120] Through the present embodiment, Vt distribution scanning is performed on the specified block at the current scanning point, and direct health status data of the memory cell at this moment, such as charge retention capability and read stability, can be obtained, which helps to evaluate the reliability of the data and whether the memory cell suffers from read disturbance, programming failure, and other faults in a timely manner. In the multi-level SSD structure, the scanning can be initiated directly at the Block level, which means that the specific location of the faulty memory cell can be accurately located.

[0121] In one exemplary embodiment, the state monitoring method of the solid state disk further includes:

[0122] The voltage threshold curve corresponding to the specified block storage at the current scan point is superimposed on the voltage threshold curve corresponding to the plurality of historical scan points of the specified block storage, and the average offset of the voltage threshold curve is determined with reference to the reference curve; the reference curve refers to the voltage threshold curve generated according to the first scan of the specified block storage when the solid state disk is shipped; in the case where the cross-layer state index of the specified block storage is greater than the preset health index, or the average offset of the voltage threshold curve of the specified block storage is greater than the preset offset threshold, an alarm report is generated.

[0123] Wherein, the average offset refers to the average difference of the data point voltage values at the same voltage when comparing the Vt curve of the current scan point with the curve of the historical scan point. The average offset is a quantitative indicator for evaluating the aging degree and performance degradation of the storage unit, which helps to dynamically monitor the change of the health state of NAND.

[0124] The reference curve is the first voltage threshold curve obtained by performing Vt distribution scanning on the specified block storage in the initial state or ideal condition of the SSD. It serves as a reference baseline for comparison and analysis of subsequent scan results, helping to identify the degradation trend of the storage unit over time.

[0125] The preset health index refers to a health index threshold set in advance, which is used to determine whether the health state of the specified block storage has deteriorated to a degree that needs attention. When the HI value exceeds the preset health index, an alarm mechanism will be triggered to remind troubleshooting or take repair measures.

[0126] The preset offset threshold refers to the threshold of the average offset of the Vt curve, which is used to determine whether the read stability and data retention capability of the storage unit have changed significantly. Once the average offset exceeds this threshold, the system will generate an alarm report to indicate that the storage unit may have suffered serious performance degradation.

[0127] Optionally, the computer device retrieves the voltage threshold curve data of the specified block stored at multiple historical scanning points (up to 5 time points) and the latest voltage threshold curve data of the current scanning point from the database; in the visualization interface, the Vt curve of the current scanning point is superimposed and displayed with the selected historical scanning point curve; the reference curve generated when the SSD is shipped is taken as a reference for comparison with the superimposed curve. By calculating the voltage difference of the corresponding points of the historical scanning point curve and the current scanning point curve with the reference curve, the average offset is counted, the offset values of different points are calculated, and the maximum offset and the minimum offset are recorded. After the configured alarm events are triggered, the time is automatically recorded and the Vth scanning is triggered, such as the HI warning and the Vth warning, the HI value is monitored in real time, and it is checked whether the cross-layer state index stored in the specified block exceeds the preset health index threshold value. Once the threshold value is reached, the subsequent alarm mechanism is triggered immediately. It is analyzed whether the average offset of the historical and current Vt curves is greater than the preset offset threshold value; if the HI value stored in the specified block is out of standard or the average offset of the Vt curve is too large, the system automatically generates an alarm report. The alarm report should include the Block information, HI value, maximum offset, minimum offset, high-wear Block proportion, average temperature, SAMAT data and other key data triggered by the alarm, as well as the recommended maintenance operation, and the global health status is summarized and displayed.

[0128] Through the embodiment, the voltage threshold curves of the specified block stored at different time points (historical scanning points and current scanning points) are superimposed and displayed, and the evolution trend of the Vt distribution with time can be intuitively analyzed. This trend analysis can find signs of performance degradation of the storage unit in advance, and can accurately locate the storage unit with the most serious performance degradation, so as to make a warning before failure occurs and take preventive measures in time; when the cross-layer state index stored in the specified block exceeds the preset threshold value or the average offset of the Vt curve exceeds the predetermined limit value, the system automatically generates an alarm report. This automatic monitoring and alarm mechanism greatly improves the efficiency of SSD maintenance and reduces the burden of manual monitoring.

[0129] In one exemplary embodiment, in order to more accurately determine the cross-layer state index of the block storage, the state monitoring of the above-mentioned solid state disk further includes the following steps after determining the cross-layer state index of the specified block storage based on multiple source data:

[0130] The computer device obtains the voltage threshold curve corresponding to the current scanning point of the specified block storage, superimposes and compares the reference curve and the voltage threshold curve corresponding to the current scanning point of the specified block storage, and obtains the real-time offset of the voltage threshold curve corresponding to the current scanning point of the specified block storage and the voltage threshold distribution change rate of the voltage threshold distribution parameter. The computer device determines the voltage state index of the specified block storage according to the real-time offset of the voltage threshold curve corresponding to the current scanning point of the specified block storage and the voltage threshold distribution change rate of the voltage threshold distribution parameter. The computer device fuses the cross-layer state index of the specified block storage and the voltage state index of the specified block storage to obtain the final cross-layer state index of the specified block storage.

[0131] The real-time offset of the voltage threshold curve corresponding to the current scanning point of the specified block storage refers to the offset between the voltage threshold curve corresponding to the current scanning point of the specified block storage and the reference curve.

[0132] The voltage threshold distribution parameter refers to a mathematical index describing the characteristics of the voltage threshold distribution (Vth distribution), including but not limited to the mean, standard deviation, distribution width, or distribution shape change of the distribution. These parameters can reflect the concentration trend and dispersion degree of the Vth distribution of the storage unit.

[0133] The voltage threshold distribution change rate refers to the change rate of a specific Vth distribution parameter within a certain time interval, which is used to quantify the change trend of the Vth distribution stability of the storage unit over time. For example, in the continuous two scans (1 hour interval), the mean value of the Vth distribution of the storage unit changes from 1.2V to 1.3V, and the change rate is (1.3V-1.2V) / 1.2V=0.0833 (about 8.33%), indicating that the Vth distribution is gradually shifting over time.

[0134] Optionally, after the computer device determines the cross-layer state index of the specified block storage based on the multi-source data, the computer device obtains the voltage threshold curve corresponding to the current scanning point of the specified block storage, superimposes and compares the reference curve and the voltage threshold curve corresponding to the current scanning point of the specified block storage, and obtains the real-time offset of the voltage threshold curve corresponding to the current scanning point of the specified block storage and the voltage threshold distribution change rate of the voltage threshold distribution parameter. The computer device determines the voltage state index of the specified block storage according to the real-time offset of the voltage threshold curve corresponding to the current scanning point of the specified block storage and the voltage threshold distribution change rate of the voltage threshold distribution parameter. The computer device fuses the cross-layer state index of the specified block storage and the voltage state index of the specified block storage to obtain the final cross-layer state index of the specified block storage.

[0135] Through the embodiment, the fusion of the cross-layer state index of the specified block storage and the voltage state index of the specified block storage cooperatively considers the Vth distribution stability of the physical layer and the cross-layer environmental influence (such as temperature, wear, and the like), so that the health index not only reflects the instant state of the storage unit, but also comprehensively considers the long-term and multi-dimensional performance change, provides more comprehensive health evaluation; the calculation of the real-time offset and the distribution change rate can accurately locate the specific reason for the performance decline of the storage unit, such as the drift of the voltage threshold with time or the too large distribution width, provides direct evidence for fault diagnosis, and accelerates the maintenance response speed.

[0136] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment.

[0137] The embodiments of the present application also provide a state monitoring device of a solid state disk, as shown in Figure 6 The device comprises:

[0138] The cross-layer acquisition module 602 is configured to acquire multi-source data of NAND flash in the solid state disk to be tested; the NAND flash comprises a plurality of block storages; the multi-source data comprises current erase-write times, average read-write error rates and real-time temperatures of the plurality of block storages;

[0139] The state prediction module 604 is configured to respectively perform weighted summation on the current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages, to obtain cross-layer state indexes of the block storages in the plurality of block storages.

[0140] The state monitoring module 606 is configured to determine a state index of the solid state disk to be tested according to the cross-layer state indexes of the block storages in the plurality of block storages, and perform state monitoring on the solid state disk to be tested based on the determined state index of the solid state disk to be tested.

[0141] In an example, the state prediction module 604 is further configured to calculate a ratio between the current erase-write times of the block storage and the maximum tolerable erase-write times of the block storage in the plurality of block storages respectively, to obtain a first ratio corresponding to the block storage in the plurality of block storages; calculate a ratio between the average read-write error rates of the block storage and the maximum read-write error rates of the block storage in the plurality of block storages respectively, to obtain a second ratio corresponding to the block storage in the plurality of block storages; calculate a temperature difference between the real-time temperature of the block storage and the optimal working temperature of the block storage in the plurality of block storages respectively, and calculate a ratio between the temperature difference of the block storage and the maximum allowed temperature of the block storage in the plurality of block storages respectively, to obtain a third ratio corresponding to the block storage in the plurality of block storages; and perform weighted summation on the first ratio corresponding to the block storage in the plurality of block storages, the second ratio corresponding to the block storage in the plurality of block storages, and the third ratio corresponding to the block storage in the plurality of block storages, to obtain a cross-layer state index of the block storage in the plurality of block storages.

[0142] In an example, the state prediction module 604 is further configured to perform the following weight adjustment operation on the block storage in the plurality of block storages as the current block storage: in a case where the real-time temperature of the current block storage is greater than a preset temperature threshold, adjusting an initial weight of the third ratio corresponding to the current block storage to obtain a first weight; the first weight is positively correlated with the real-time temperature of the current block storage and is less than a preset weight threshold; adjusting an initial weight of the second ratio corresponding to the current block storage according to the first weight to obtain a second weight, and adjusting an initial weight of the first ratio corresponding to the current block storage according to the first weight to obtain a third weight; the sum of the first weight, the second weight, and the third weight is equal to 1; and the weighted summation on the first ratio corresponding to the block storage in the plurality of block storages, the second ratio corresponding to the block storage in the plurality of block storages, and the third ratio corresponding to the block storage in the plurality of block storages is performed according to the first weight, the second weight, and the third weight.

[0143] In an example, the state prediction module 604 is further configured to, in a case where the real-time temperature of the current block storage is greater than a preset temperature threshold, determine a temperature difference between the real-time temperature of the current block storage and the maximum allowed temperature of the current block storage as a real-time temperature difference, and determine a ratio between the real-time temperature difference and a preset weight influence coefficient as a temperature change proportion; determine a first weight adjustment factor according to the temperature change proportion, and determine a product between the first weight adjustment factor and an initial weight of the third ratio corresponding to the current block storage as the first weight.

[0144] In an example, the state prediction module 604 is further configured to determine a current remaining proportion of the first weight with respect to the specified weight state according to the first weight, and determine an original remaining proportion of the initial weight of the third ratio corresponding to the current block storage with respect to the specified weight state according to the current block storage storing the initial weight of the third ratio, determine a second weight adjustment factor as a ratio between the current remaining proportion and the original remaining proportion, and determine the second weight as a product of the second weight adjustment factor and the initial weight of the second ratio corresponding to the current block storage.

[0145] In an example, the device further comprises a visualization module configured to construct a multi-level addressing heat map according to a physical structure of the solid state drive under test; the multi-level addressing heat map is a multi-level view constructed by taking the cross-layer state index of the block storage as a heat value; the multi-level addressing heat map comprises a channel level view, a target level view, a logical unit level view, and a block level view; the channel level view shows a health state summary of a channel level in the solid state drive under test; the target level view shows a health state summary of a target level in the solid state drive under test; the logical unit level view shows the cross-layer state index of the block storage under a logical unit level in the solid state drive under test; the block level view shows attributes of a word line under a block level in the solid state drive under test; the word line is a control line for selecting a specific storage unit; the multi-level addressing heat map is used to locate and display a set of target block storages; the set of target block storages refers to at least one block storage with a cross-layer state index greater than a preset health index.

[0146] In an example, the visualization module is further configured to initiate a scan instruction package for a specified block storage within the block level view; the scan instruction package comprises a voltage threshold scan configuration parameter; the scan instruction package is used to instruct the solid state drive under test to cyclically scan the specified block storage according to the voltage threshold scan configuration at a current scan point and return scan data corresponding to the current scan point; draw a voltage threshold curve corresponding to the current scan point of the specified block storage according to the scan data; the voltage threshold curve corresponding to the current scan point is used to analyze the read state and data reliability of the NAND flash memory at the current scan point.

[0147] In an example, the visualization module is further configured to select voltage threshold curves corresponding to a plurality of historical scan points of the specified block storage, superimpose the voltage threshold curve corresponding to the current scan point of the specified block storage and the voltage threshold curves corresponding to the plurality of historical scan points, and determine an average offset of the voltage threshold curve with reference to a reference curve; the reference curve refers to a voltage threshold curve generated by scanning the specified block storage for the first time when the solid state drive under test is shipped; generate an alarm report in a case where the cross-layer state index of the specified block storage is greater than a preset health index, or the average offset of the voltage threshold curve of the specified block storage is greater than a preset offset threshold.

[0148] The features of the embodiments of the state monitoring device of the solid state disk can be referred to the related descriptions of the embodiments of the state monitoring method of the solid state disk, which will not be repeated here.

[0149] The embodiments of the present application also provide an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor is configured to execute the computer program to perform the steps in any of the embodiments of the state monitoring method of the solid state disk.

[0150] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to perform the steps in any of the embodiments of the state monitoring method of the solid state disk when executed.

[0151] In an example embodiment, the computer readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0152] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps in any of the embodiments of the state monitoring method of the solid state disk.

[0153] The embodiments of the present application also provide another computer program product, which comprises a non-volatile computer readable storage medium, and the non-volatile computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in any of the embodiments of the state monitoring method of the solid state disk.

[0154] The skilled person can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0155] The above describes in detail the state monitoring method of the solid state disk, the electronic device and the storage medium provided by the application. The principles and implementation manners of the application are described by using specific examples in this paper, and the above description of the embodiments is only used to help understand the method of the application and its core idea. It should be pointed out that, for ordinary skilled persons in the technical field, some improvements and modifications can be made to the application without departing from the principles of the application, and these improvements and modifications also fall within the protection scope of the claims of the application.

Claims

1. A state monitoring method of a solid state drive, characterized by, The method comprises: obtaining multi-source data of NAND flash in a to-be-tested solid state disk; the NAND flash comprises a plurality of block storages; the multi-source data comprises current erase-write times, average read-write error rates and real-time temperatures of the plurality of block storages; respectively performing weighted summation on the current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages to obtain cross-layer state indexes of the block storages in the plurality of block storages; constructing a multi-level addressing thermal map according to a physical structure of the to-be-tested solid state disk; locating and displaying a group of target block storages through the multi-level addressing thermal map; the group of target block storages refers to at least one block storage whose cross-layer state index is greater than a preset health index; wherein the multi-level addressing thermal map is a multi-level view constructed by taking the cross-layer state indexes of the block storages as thermal values; the multi-level addressing thermal map comprises a channel level view, a target level view, a logical unit level view and a block level view; the channel level view displays a health status summary of a channel level in the to-be-tested solid state disk; the target level view displays a health status summary of a target level in the to-be-tested solid state disk; the logical unit level view displays the cross-layer state indexes of the block storages under a logical unit level in the to-be-tested solid state disk; and the block level view displays attributes of word lines under a block level in the to-be-tested solid state disk; the word line is a control line for selecting a storage unit.

2. The method of claim 1, wherein, The method further comprises: respectively performing weighted summation on the current erase-write times, the average read-write error rates and the real-time temperatures of the block storages in the plurality of block storages to obtain cross-layer state indexes of the block storages in the plurality of block storages, comprises: respectively calculating ratios between the current erase-write times of the block storages in the plurality of block storages and maximum resistant erase-write times of the block storages in the plurality of block storages to obtain corresponding first ratios of the block storages in the plurality of block storages; respectively calculating ratios between the average read-write error rates of the block storages in the plurality of block storages and maximum read-write error rates of the block storages in the plurality of block storages to obtain corresponding second ratios of the block storages in the plurality of block storages; respectively calculating temperature difference values between the real-time temperatures of the block storages in the plurality of block storages and optimal working temperatures of the block storages in the plurality of block storages, and respectively calculating ratios between the temperature difference values of the block storages in the plurality of block storages and maximum allowed temperatures of the block storages in the plurality of block storages to obtain corresponding third ratios of the block storages in the plurality of block storages; 3. The method of claim 2, wherein, performing weighted summation on the corresponding first ratios of the block storages in the plurality of block storages, the corresponding second ratios of the block storages in the plurality of block storages and the corresponding third ratios of the block storages in the plurality of block storages to obtain the cross-layer state indexes of the block storages in the plurality of block storages. The method further comprises: performing the following weight adjustment operation on the block storages in the plurality of block storages as current block storages respectively: adjusting the initial weight of the third ratio corresponding to the current block storage to obtain a first weight in a case where the real-time temperature of the current block storage is greater than a preset temperature threshold; the first weight is positively correlated with the real-time temperature of the current block storage and is less than a preset weight threshold; adjusting the initial weight of the second ratio corresponding to the current block storage according to the first weight to obtain a second weight, and adjusting the initial weight of the first ratio corresponding to the current block storage according to the first weight to obtain a third weight; the sum of the first weight, the second weight and the third weight is equal to 1; wherein the weighted summation of the first ratio corresponding to the block storage in the plurality of block storages, the second ratio corresponding to the block storage in the plurality of block storages and the third ratio corresponding to the block storage in the plurality of block storages is performed according to the first weight, the second weight and the third weight.

4. The method of claim 3, wherein, The adjusting the initial weight of the third ratio corresponding to the current block storage to obtain a first weight in a case where the real-time temperature of the current block storage is greater than a preset temperature threshold, comprises: In a case where the real-time temperature of the current block storage is greater than a preset temperature threshold, the temperature difference between the real-time temperature of the current block storage and the maximum allowable temperature of the current block storage is determined as a real-time temperature difference, and the ratio between the real-time temperature difference and a preset weight influence coefficient is determined as a temperature change proportion; determining a first weight adjustment factor according to the temperature change proportion, and determining the product between the first weight adjustment factor and the initial weight of the third ratio corresponding to the current block storage as the first weight.

5. The method of claim 3, wherein, The adjusting the initial weight of the second ratio corresponding to the current block storage according to the first weight to obtain a second weight, comprises: determining a current remaining proportion of the first weight with respect to a specified weight state according to the first weight, and determining an original remaining proportion of the initial weight of the third ratio corresponding to the current block storage with respect to the specified weight state according to the initial weight of the third ratio corresponding to the current block storage; determining a second weight adjustment factor as the ratio between the current remaining proportion and the original remaining proportion, and determining the product between the second weight adjustment factor and the initial weight of the second ratio corresponding to the current block storage as the second weight.

6. The method of claim 1, wherein, The method further comprises: initiating a scan instruction package for a specified block storage within the block level view; the scan instruction package comprises a voltage threshold scan configuration parameter; the scan instruction package is used to instruct the to-be-tested solid state disk to cyclically scan the specified block storage according to the voltage threshold scan configuration at a current scan point and return scan data corresponding to the current scan point; drawing a voltage threshold curve corresponding to the current scan point of the specified block storage according to the scan data; the voltage threshold curve corresponding to the current scan point is used to analyze the read state and data reliability of the NAND flash at the current scan point.

7. The method of claim 6, wherein, The method further comprises: The voltage threshold curve corresponding to the plurality of historical scanning points of the specified block storage is selected, the voltage threshold curve corresponding to the current scanning point of the specified block storage is superimposed on the voltage threshold curve corresponding to the plurality of historical scanning points, and the average offset of the voltage threshold curve is determined with reference to a reference curve; the reference curve refers to the voltage threshold curve generated according to the first scanning of the specified block storage when the to-be-tested solid state disk is shipped. In the case where the cross-layer state index of the specified block storage is greater than the preset health index, or the average offset of the voltage threshold curve of the specified block storage is greater than the preset offset threshold, an alarm report is generated.

8. An electronic device, comprising: Comprise: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the state monitoring method of the solid state disk according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, wherein the computer program is executed by the processor to implement the steps of the state monitoring method of the solid state disk according to any one of claims 1 to 7.

Images