CN118312109A - Bad block management method, system, medium and product of industrial solid state disk - Google Patents
Bad block management method, system, medium and product of industrial solid state disk Download PDFInfo
- Publication number
- CN118312109A CN118312109A CN202410734289.3A CN202410734289A CN118312109A CN 118312109 A CN118312109 A CN 118312109A CN 202410734289 A CN202410734289 A CN 202410734289A CN 118312109 A CN118312109 A CN 118312109A
- Authority
- CN
- China
- Prior art keywords
- preset
- storage unit
- data
- solid state
- bad block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000007726 management method Methods 0.000 title claims abstract description 99
- 239000007787 solid Substances 0.000 title claims abstract description 90
- 230000036541 health Effects 0.000 claims abstract description 76
- 238000000034 method Methods 0.000 claims abstract description 56
- 238000013508 migration Methods 0.000 claims abstract description 24
- 230000005012 migration Effects 0.000 claims abstract description 24
- 230000000694 effects Effects 0.000 claims abstract description 15
- 230000005856 abnormality Effects 0.000 claims abstract description 11
- 238000013210 evaluation model Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 18
- 238000001514 detection method Methods 0.000 claims description 16
- 238000002955 isolation Methods 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000011157 data evaluation Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 14
- 238000012545 processing Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 11
- 238000012795 verification Methods 0.000 description 10
- 230000004044 response Effects 0.000 description 9
- 210000004027 cell Anatomy 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 7
- 230000035945 sensitivity Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Landscapes
- Debugging And Monitoring (AREA)
Abstract
A bad block management method, system, medium and product of industrial solid state disk relate to the field of solid state disk management, the method includes: obtaining a surface temperature set corresponding to each storage unit respectively; determining a storage unit meeting a preset abnormality judgment rule as a storage unit to be detected; the method comprises the steps of inputting the writing times, the writing intensity and the error rate of a storage unit to be detected in a preset duration into a preset health evaluation model to obtain the health index of the storage unit to be detected; determining a storage unit to be detected as an early warning block, backing up effective data in the early warning block and executing data migration operation; and determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block and isolating the bad block from the industrial solid state disk. By implementing the method, the industrial solid state disk is actively, accurately and comprehensively monitored in health state, faults are early warned and positioned, and important data can be protected.
Description
Technical Field
The application relates to the field of solid state disk management, in particular to a bad block management method, a bad block management system, a bad block management medium and a bad block management product of an industrial solid state disk.
Background
With the rapid development of data centers and industrial automation, the requirements on the performance and the reliability of storage devices are higher and higher. Industrial Solid State Disk (SSD) has become the storage device of choice for many critical applications because of its high-speed data processing capability and high shock resistance. However, the solid state disk may have a bad block problem in the long-term operation process, which forms a serious threat to the security of data and the stable operation of the system. Therefore, how to effectively manage bad blocks in the solid state disk so as to improve the reliability and efficiency of operation of the solid state disk becomes an important research direction.
In the related art, a tool provided by a manufacturer of the solid state disk is generally used to detect the solid state disk, so as to scan the solid state disk and detect a bad block. The S.M.A.R.T. information of the magnetic disk can also be checked, and the S.M.A.R.T. information of the solid state disk can be obtained through an operating system or third party software, so as to judge whether a bad block exists.
Although the use of manufacturer tools and s.m. a.r.t. information to detect bad blocks of Solid State Drives (SSDs) is a currently commonly employed method, passive detection by the controller firmware cannot be used for active and accurate health status monitoring of the solid state drives, and failure early warning and localization cannot be performed.
Disclosure of Invention
The application provides a bad block management method, a system, a medium and a product of an industrial solid state disk, which are used for actively and accurately monitoring the health state of the industrial solid state disk, early warning and positioning faults and protecting important data.
In a first aspect, the present application provides a method for managing bad blocks of an industrial solid state disk, which is applied to a management system, and the method includes: periodically acquiring the surface temperature of each storage unit on the industrial solid state disk to obtain a surface temperature set corresponding to each storage unit; determining a storage unit meeting a preset abnormality judgment rule as a storage unit to be detected, wherein the preset abnormality judgment rule is that the values of the surface temperatures exceeding a preset number in the surface temperature set are not within a preset working temperature threshold; the method comprises the steps of inputting the writing times, the writing intensity and the error rate of a storage unit to be detected in a preset duration into a preset health evaluation model to obtain the health index of the storage unit to be detected; when the health index is lower than a preset first threshold value and higher than a preset second threshold value, determining the storage unit to be detected as an early warning block, backing up effective data in the early warning block, and executing data migration operation, wherein the data migration operation is used for migrating the effective data from the early warning block to other storage units; and when the health index is lower than a preset second threshold value, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block, and isolating the bad block from the industrial solid state disk, wherein the preset first threshold value is larger than the preset second threshold value.
In the above embodiment, the management system first periodically obtains the surface temperature of each storage unit on the industrial solid state disk, determines the storage unit to be detected according to the abnormal temperature condition, then obtains the health index according to the writing times, writing strength and error rate of the storage unit to be detected, and divides the storage unit to be detected into the early warning block and the bad block according to different thresholds. The early warning blocks are subjected to data backup and data migration, and bad blocks are isolated, so that active monitoring and scientific management of the industrial solid state disk are realized, fault signs are found timely, early warning response is made, overall performance reduction or damage caused by the increase of the number of the bad blocks in the industrial solid state disk is effectively avoided, and long-term reliable operation of the solid state disk in an industrial environment is ensured.
In combination with some embodiments of the first aspect, in some embodiments, when the health index is lower than a preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block and isolating the bad block from the industrial solid state disk, specifically including: when the health index is lower than a preset second threshold value, determining the storage unit to be detected as a bad block; searching a logic address corresponding to the physical address of the bad block according to the physical address of the bad block; marking the physical address of the bad block as forbidden, and completing the operation of isolating the bad block from the industrial solid state disk; marking the data corresponding to the logical address as damaged data to stop the read-write activity of the bad block.
In the above embodiment, a specific technical flow of performing isolation after the bad block is determined is further described, that is, by searching the corresponding relationship between the physical address and the logical address of the bad block, the physical address of the bad block is marked as disabled, and the data corresponding to the logical address of the bad block is marked as damaged data, thereby stopping the read-write operation of the bad block. The isolation of the bad blocks is realized, the access to the bad blocks during data reading and writing is avoided, and the data safety and the stability of the industrial solid state disk are improved.
With reference to some embodiments of the first aspect, in some embodiments, before the step of periodically obtaining the surface temperatures of each storage unit on the industrial solid state disk to obtain the surface temperature sets corresponding to each storage unit respectively, the method further includes: obtaining the predicted residual life of each storage unit on the industrial solid state disk; and if the predicted remaining life is lower than the preset life threshold, determining the storage unit with the predicted remaining life lower than the preset life threshold as a bad block, and isolating the bad block from the industrial solid state disk.
In the above embodiment, the management system determines in advance whether the predicted remaining lifetime of the storage unit is lower than the preset lifetime threshold, and if so, indicates that the storage unit has not satisfied the usage requirement, and may directly determine that the bad block is isolated. The management of the life cycle of the industrial solid state disk is realized, and the storage unit which reaches the end of the expected life is isolated in advance, so that the health state of the industrial solid state disk is monitored more comprehensively, and the long-term stable and reliable operation of the industrial solid state disk is better ensured.
In combination with some embodiments of the first aspect, in some embodiments, periodically obtaining surface temperatures of each storage unit on the industrial solid state disk to obtain a surface temperature set corresponding to each storage unit, where the method specifically includes: if the predicted remaining life is higher than the preset life threshold, acquiring the surface temperature of each storage unit on the industrial solid state disk at a plurality of preset time points; and arranging the surface temperatures of the storage units according to a time sequence to obtain a surface temperature set corresponding to each storage unit.
In the above-described embodiment, the technical details of acquiring the surface temperature set of the memory cell are clarified, and for the memory cell whose predicted remaining lifetime is higher than the preset lifetime threshold, the surface temperature is acquired at a plurality of preset time points and arranged in time series as the temperature set of the memory cell. The storage unit which reaches the end of the service life is filtered, the temperature change of the normal storage unit is monitored more accurately, and reliable basic data is provided for subsequent abnormal temperature judgment.
With reference to some embodiments of the first aspect, in some embodiments, before the step of periodically obtaining the surface temperatures of each storage unit on the industrial solid state disk to obtain the surface temperature sets corresponding to each storage unit respectively, the method further includes: acquiring storage data respectively corresponding to each storage unit stored on the industrial solid state disk; extracting a plurality of characteristic information of the stored data, wherein the characteristic information comprises a data format, a data size, a data creation time, a data naming mode and data creator information; performing matching analysis on a plurality of characteristic information based on a preset information base, and determining the security level of stored data; if the matching analysis fails, inputting the plurality of characteristic information into a preset data evaluation model to obtain a security level score and a corresponding security level.
In the embodiment, the security evaluation flow of the stored data corresponding to the storage unit is increased, the security level of the stored data corresponding to different storage units can be obtained by extracting the characteristic information of the stored data and matching and judging the security level, a basis is provided for subsequent early warning response and bad block isolation, the processing priority based on the data security level is realized, and the data security is improved.
With reference to some embodiments of the first aspect, in some embodiments, the method further includes: determining the data detection frequency and the data migration rate of the early warning block according to the security levels of the stored data respectively corresponding to the storage units; and determining the isolation priority of the bad blocks according to the security levels of the storage data respectively corresponding to the storage units.
In the embodiment, the processing strategies of the early warning block and the bad block are determined according to the data security level, namely the data detection frequency of the early warning block, the data migration rate and the isolation priority of the bad block, so that higher monitoring granularity and faster response speed are given to the storage unit of important data, the security level of the stored data directly influences the processing strategies, and the security guarantee of the important data is further improved.
In combination with some embodiments of the first aspect, in some embodiments, before the step of inputting the number of writing times, the writing strength and the error rate of the storage unit to be detected in the preset duration into the preset health assessment model, the method further includes: acquiring the historical writing times, the historical writing intensity, the historical error rate and the historical health index of the training storage unit in the historical preset time; marking the historical writing times, the historical writing strength and the historical error rate as input features, and marking the historical health index as output features; training the preset model by using the input characteristics and the output characteristics to obtain the accuracy of the preset model; and when the accuracy exceeds a preset accuracy threshold, obtaining a health assessment model.
In the above embodiment, the training process of generating the health evaluation model is described, and the health evaluation model enables the evaluation of the health index to be more accurate and reliable, so that a solid foundation is provided for subsequent early warning judgment and isolation processing.
In a second aspect, an embodiment of the present application provides a management system, including: one or more processors and memory; the memory is coupled to the one or more processors, the memory for storing computer program code comprising computer instructions that the one or more processors call for causing the management system to perform the method as described in the first aspect and any one of the possible implementations of the first aspect.
In a third aspect, embodiments of the present application provide a computer program product comprising instructions which, when run on a management system, cause the management system to perform a method as described in the first aspect and any possible implementation manner of the first aspect.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium comprising instructions which, when run on a management system, cause the management system to perform a method as described in the first aspect and any possible implementation manner of the first aspect.
It will be appreciated that the management system provided in the second aspect, the computer program product provided in the third aspect and the computer storage medium provided in the fourth aspect are each configured to perform the method provided by the embodiment of the present application. Therefore, the advantages achieved by the method can be referred to as the advantages of the corresponding method, and will not be described herein.
One or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
1. The technical scheme of comprehensively evaluating and judging the health state of the storage unit aiming at the surface temperature, the writing times, the writing intensity and the error rate of the industrial solid state disk is adopted, so that the active, dynamic and accurate health monitoring of the industrial solid state disk can be realized, the problem that the solid state disk failure cannot be early-warned and accurately positioned by means of passive detection and single sampling detection of the controller firmware in the related technology is effectively solved, and the effects of greatly reducing the solid state disk failure rate and remarkably improving the operation reliability in the industrial environment are further realized.
2. The processing strategies of the early warning block and the bad block are determined according to the data security level, namely the data detection frequency of the early warning block, the data migration rate and the isolation priority of the bad block, so that higher monitoring granularity and faster response speed are given to the storage unit of important data, the security level of the stored data directly influences the processing strategies, and the security guarantee of the important data is further improved.
Drawings
FIG. 1 is a schematic flow chart of a bad block management method of an industrial solid state disk according to an embodiment of the present application;
FIG. 2 is another flow chart of a bad block management method of an industrial solid state disk according to an embodiment of the present application;
fig. 3 is a schematic diagram of a physical device of a management system according to an embodiment of the application.
Detailed Description
The terminology used in the following embodiments of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates to the contrary. It should also be understood that the term "and/or" as used in this disclosure is intended to encompass any or all possible combinations of one or more of the listed items.
The terms "first," "second," and the like, are used below for descriptive purposes only and are not to be construed as implying or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature, and in the description of embodiments of the application, unless otherwise indicated, the meaning of "a plurality" is two or more.
Fig. 1 is a schematic flow chart of a bad block management method of an industrial solid state disk according to an embodiment of the application.
S101, periodically acquiring the surface temperature of each storage unit on the industrial solid state disk to obtain a surface temperature set corresponding to each storage unit;
The management system may set a fixed time interval, such as acquiring the surface temperature of all storage units every 10 minutes, every half hour, or every hour. The sensor for acquiring the surface temperature can be integrated in the industrial solid state disk, or can be externally connected to the surface of the industrial solid state disk, and the sensor is not limited herein. The sensing accuracy of the surface temperature can reach 0.1 ℃. The sensed temperature data is fed back to the management system in real time.
For example, if the industrial solid state disk has 3 storage units A, B, C, the surface temperature obtained by the management system each time will be 3 temperature values. If the management system obtains the surface temperature of each storage unit on the industrial solid state disk every 10 minutes, and the surface temperature of each storage unit on the industrial solid state disk is obtained from 1:00 to 2:00, in the hour, the management system obtains that the surface temperature of A is 12 ℃, the surface temperature of B is 17 ℃ and the surface temperature of C is 52 ℃ at 1:00; the management system obtains a surface temperature of 18 ℃ for A, 21 ℃ for B and 30 ℃ for C at 1:10; the management system obtains a surface temperature of 72 ℃ for A, 57 ℃ for B and 50 ℃ for C at 1:20; the management system obtains a surface temperature of 67 ℃ for A, 38 ℃ for B, and 79 ℃ for C at 1:30; the management system obtains that the surface temperature of A is 77 ℃, the surface temperature of B is 35 ℃ and the surface temperature of C is 70 ℃ at 1:40; the management system obtains that the surface temperature of A is 75 ℃, the surface temperature of B is 58 ℃ and the surface temperature of C is 69 ℃ at 1:50; the management system obtains a surface temperature of 80C, B50C, C60℃ at 2:00. The obtained surface temperature sets respectively corresponding to the storage units are as follows: the surface temperature set of A is {12 ℃,18 ℃, 72 ℃, 67 ℃,77 ℃, 75 ℃,80 ℃, the surface temperature set of B is {17 ℃,21 ℃,57 ℃, 38 ℃, 35 ℃, 38 ℃,50 ℃, and the surface temperature set of C is {52 ℃,30 ℃,50 ℃, 70 ℃, 69 ℃,60 ℃.
S102, determining a storage unit meeting a preset abnormality judgment rule as a storage unit to be detected, wherein the preset abnormality judgment rule is that the values of the surface temperatures exceeding the preset number in the surface temperature set are not within a preset working temperature threshold;
And the management system determines the storage units meeting the preset abnormality judgment rules as the storage units to be detected so as to carry out subsequent operation. The preset abnormality determination rule here is that the value of the surface temperature exceeding the preset number of surface temperatures in the surface temperature set is not within the preset operating temperature threshold. That is, if the number of values of the surface temperature in the set of surface temperatures of a certain memory cell that are not within the preset operating temperature threshold exceeds the preset number, the memory cell will be determined as the memory cell to be detected.
Assuming that the preset operating temperature threshold is 10-70 degrees and the preset number is 4, the storage unit a is to be regarded as the storage unit to be detected, taking the example of the surface temperature set A, B, C in step S101.
S103, inputting the writing times, the writing strength and the error rate of the storage unit to be detected in a preset time period into a preset health evaluation model to obtain the health index of the storage unit to be detected;
The management system inputs the writing times, the writing intensity and the error rate of the storage unit to be detected in the preset time period into a preset health assessment model to obtain the health index of the storage unit to be detected. The writing times are used for counting the total reading and writing times of the storage unit in a preset time, the writing strength is used for calculating the reading and writing times in unit time, and the error rate is used for counting errors or retry times in the reading and writing process. The above parameters reflect the usage and health of the memory unit. The writing times, writing intensity and error rate of the storage unit to be detected in the preset duration are input into a preset health evaluation model, so that a health index with a value of 0-100 can be obtained, and the larger the number is, the better the health condition is.
The method for constructing the health assessment model based on deep learning comprises the following steps:
First, the management system collects, from the storage system, the number of historical writes, the strength of the historical writes, and the historical error rate of a plurality of storage units to be detected over a predetermined period of time in the past (e.g., over 6 months). The historical writing times are the total number of the reading and writing requests of each day in the past preset time period of the storage unit to be detected, which is collected by the management system. The historical writing strength is the number of read-write requests per second in the past preset time length of the storage unit to be detected, which is collected by the management system. The historical error rate is the ratio of the number of errors in the reading and writing process of the storage unit to be detected, which is acquired by the management system, in the past preset time length to the total request number. In addition, the management system records the historical health index of each storage unit to be detected in the past preset time period, and the historical health index of each storage unit to be detected can be obtained through historical maintenance record or expert evaluation, which is not limited herein. The management system stores the collected historical writing times, the historical writing intensity, the historical error rate and the historical health index of each storage unit to be detected into a data set D according to the time sequence, wherein each data format is (the historical writing times, the historical writing intensity, the historical error rate and the historical health index). The historical writing times, the historical writing intensity and the historical error rate are input features of model training, and the historical health index is output features of model training.
Secondly, the management system carries out data preprocessing on the data set D, deletes missing data and abnormal data (such as records with error rate of more than 1) in the data set D, and normalizes three characteristics of historical writing times, historical writing intensity and historical error rate to be between 0 and 1.
The management system then builds a LSTM based recurrent neural network comprising an input layer, 2 LSTM hidden layers, a fully connected layer and an output layer. The input layer inputs the normalized historical writing times, the historical writing intensity and the historical error rate, the hidden layer node number is set to 64, and the full connection layer node number is set to 32. The output layer is a Sigmoid activation function, and a probability value of 0-1 can be obtained to represent the fault risk. Then, the probability value of 0-1 is converted into a health index with the value of 0-100 as output. The Sigmoid function is a commonly used activation function, and its mathematical expression is f (x) =1/(1+e-x), and the shape is an S-shaped curve. The Sigmoid function compresses the input value between 0 and 1, and is widely used in neural networks to introduce nonlinear characteristics and limit the output of neurons.
Next, the management system adopts Adam optimizer, the learning rate is set to 0.001, the training batch size is set to 32, and the training batch size can be set according to the actual situation, which is not limited herein. 80% of the historical data is divided into training sets, 20% is divided into verification sets, 100 epochs are trained, and a model with the highest verification set accuracy is stored. Epoch is the process by which the training dataset passes through the neural network completely once. In machine learning and deep learning, epoch is a unit used to measure the number of times the entire training set is repeatedly learned. Specifically, when the neural network completes one forward computation and back propagation process, i.e., all data has been processed once by the network, one epoch is completed. The management system uses the two kinds of cross entropy as a loss function, early Stopping is adopted to prevent overfitting, and when the value of the loss function exceeds a preset function threshold, model training is confirmed to be completed, so that a health assessment model is obtained. Early Stopping is a technique in deep learning and machine learning that prevents model overfitting by monitoring the performance of the model on the validation set to determine when to stop training.
Finally, the management system inputs the input features in the verification set to the health assessment model, then obtains the prediction output of the health assessment model, compares the prediction output of the health assessment model with the actual output features in the verification set, and uses some performance indexes such as accuracy, precision, recall, F1 score, mean Square Error (MSE) and the like to assess the performance of the health assessment model. According to the performance of the health assessment model on the verification set, parameters of the health assessment model are adjusted, including adjusting learning rate, changing model complexity (such as increasing or decreasing the number of layers or nodes of the neural network), modifying regularization strength, and the like. This process may require multiple iterations, each of which is adjusted based on previous learning results, to optimize the health assessment model.
S104, when the health index is lower than a preset first threshold value and higher than a preset second threshold value, determining the storage unit to be detected as an early warning block, backing up effective data in the early warning block, and executing data migration operation, wherein the data migration operation is used for migrating the effective data from the early warning block to other storage units;
When the health index is lower than a preset first threshold value and higher than a preset second threshold value, the management system determines the storage unit to be detected as an early warning block, backs up effective data in the early warning block, and executes data migration operation, wherein the data migration operation is used for migrating the effective data from the early warning block to other storage units. Assuming that the preset first threshold is 70 and the preset second threshold is 40, when the health index of a memory cell is 60, the memory cell is determined as an early warning block. For the early warning block, the management system starts an early warning response mechanism, backs up the effective data in the early warning block and avoids data loss. After the backup is completed, the management system migrates the effective data in the early warning block to other relatively healthy storage units in the industrial solid state disk. The migration mode can adopt different strategies according to different security levels corresponding to the stored data, so that the rapid migration of the important data is ensured.
And S105, when the health index is lower than a preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block, and isolating the bad block from the industrial solid state disk, wherein the preset first threshold is larger than the preset second threshold.
And when the health index is lower than a preset second threshold value, the management system determines the storage unit to be detected as a bad block, stops the read-write activity of the bad block and isolates the bad block from the industrial solid state disk. With the example in step S104, the storage unit to be detected with a health index of 30 is confirmed as a bad block. For the storage unit detected as the bad block, the management system stops all read-write operations of the bad block, and isolates the bad block from other normal storage units in the industrial solid state disk, so that interference on data access is avoided. The isolation may be to clear the logical address information of the corresponding bad block, mark its physical space as unavailable, and modify the address mapping table in the firmware to mask the bad block. When the number of bad blocks exceeds the redundancy tolerance of the industrial solid state disk, the management system should report to the user that the solid state disk needs to be replaced.
Optionally, when the health index is lower than a preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block and isolating the bad block from the industrial solid state disk, where the preset first threshold is greater than the preset second threshold can be implemented by the following ways: when the health index is lower than a preset second threshold value, determining the storage unit to be detected as a bad block; searching a logic address corresponding to the physical address of the bad block according to the physical address of the bad block; marking the physical address of the bad block as forbidden, and completing the operation of isolating the bad block from the industrial solid state disk; marking the data corresponding to the logical address as damaged data to stop the read-write activity of the bad block.
The method provided in this embodiment will be described in more detail. Fig. 2 is a schematic flow chart of a bad block management method of an industrial solid state disk according to an embodiment of the application.
S201, acquiring storage data respectively corresponding to each storage unit stored on an industrial solid state disk;
the management system can obtain the content of the storage data in each storage unit through the main scanning of the storage data by a read-write command, and can also obtain the use and file information of each storage unit from the operating system at regular intervals as the storage data.
S202, extracting a plurality of characteristic information of stored data, wherein the characteristic information comprises a data format, a data size, a data creation time, a data naming mode and data creator information;
After the management system obtains the stored data, the content of the stored data is understood according to the characteristic information. The feature information includes feature information such as data format, data size, data creation time, data naming mode, data creator information, etc.
The data format determines the manner in which the data is presented and the readability. For example, text files, picture files, audio files, video files, etc. all have respective formats. Knowing the format of the data helps determine the purpose and sensitivity of the data, e.g., files that relate to personal privacy or confidential information may have a higher sensitivity, which typically employs a particular data format.
The data size may reflect the importance and value of its content. For example, a large database file may contain a large amount of critical business data, while a smaller configuration file may contain only some system setup information. By analyzing the data size, the value and sensitivity of the data can be primarily judged.
The data creation time may provide information about the freshness and timeliness of the data. For example, recently created files may contain up-to-date business data or personal information, while files many years ago may have been out of date or no longer need frequent access.
The data naming scheme may provide clues as to the type and purpose of the data. For example, a file named in a particular format or schema may contain information about the system configuration, logging, or user data. By analyzing the data naming scheme, the sensitivity and importance of the data can be further inferred.
The data creator information may be aware that the creator of the data (e.g., user ID, software developer, etc.) may provide information about the source and reliability of the data. For example, data from a particular system administrator or software developer may have higher reliability, while data from an unknown or untrusted source may have higher risk.
Thus, by extracting and analyzing such characteristic information, the management system can understand the content of the stored data to perform sensitivity evaluation on the stored data and determine the corresponding security level.
S203, carrying out matching analysis on a plurality of characteristic information based on a preset information base, and determining the security level of the stored data;
And the management system performs matching analysis on the extracted characteristic information and a preset information base. The preset information base stores the corresponding relation between different data characteristic information and the security level. The preset information base is a database containing known characteristic information corresponding to the security level standard and is used for comparing and matching with the extracted characteristic information. The preset information base can be predefined and updated to accommodate changing threat and data security requirements. Specifically, the management system performs matching analysis on the extracted characteristic information and a preset information base, and the process is as follows:
Data format matching: and comparing the extracted data format with format standards in a preset information base to determine whether a known sensitive format or a malicious format exists.
Data size analysis: and comparing the size of the extracted data with a size threshold value in a preset information base to judge whether the data exceeds a normal range or accords with a specific safety standard.
Data creation time assessment: comparing the extracted data creation time with time standards in a preset information base to determine whether suspicious creation time patterns or timestamp anomalies exist.
Data naming method review: the extracted data naming method is analyzed whether the known normal mode is met or suspicious naming features exist. For example, a particular naming convention may be related to malware or sensitive files.
Data creator information verification: the extracted data creator information is compared with trusted sources in a pre-set information base to determine whether the source of the data is trusted or has a potential threat.
From these matching analyses, the security level of the stored data can be derived. The security level may be classified according to the severity of the matching result, such as low risk, medium risk, and high risk.
On the other hand, if the source of the stored data is reliable, the content relates to sensitive information, the use is important and the use frequency is high, the characteristic information is matched with the high security level characteristic in the preset information base, and the management system determines the security level to be high. In contrast, if the source of the stored data is unknown, the content does not relate to sensitive information, the use is unimportant and the frequency of use is low, the characteristic information matches with the low security level characteristic in the preset information base, and the management system determines the security level thereof to be low.
S204, if the matching analysis fails, inputting a plurality of characteristic information into a preset data evaluation model to obtain a security level score and a corresponding security level;
If the matching analysis fails, the matching result is empty, and the preset information base cannot judge the security level of certain stored data, namely the matching analysis fails. At this time, the management system may call a data evaluation model preset based on the artificial intelligence algorithm, directly analyze the feature information of the stored data, give a security level prediction score, i.e. a security level score, and convert the security level prediction score into a corresponding security level, e.g. score the privacy information, and determine the security level according to the score.
The training method of the data evaluation model is as follows:
First, a data set is prepared, which includes a plurality of feature information and corresponding security level scores. And, the dataset contains various different types of data so that the model can learn associations and patterns between different features; and secondly, selecting a proper open source model for training according to the characteristics of data and task requirements. Common models include decision trees, support vector machines, neural networks, etc.; the data evaluation model is then constructed using the selected model. An open-source deep learning framework such as TensorFlow, pyTorch and the like can be used, or other programming languages such as Python and Java and the like can be used, wherein super parameters are parameters which need to be manually set in the model training process, such as learning rate, batch size, iteration number and the like, and the setting of the super parameters has great influence on the training effect of the model and needs to be adjusted according to actual conditions; then training the model by using the prepared data set, setting super parameters in the training process, and adjusting according to errors of the training set and errors of the verification set; finally, the trained model is evaluated using the test dataset. The evaluation index comprises an accuracy rate, a recall rate, an F1 value and the like. And adjusting and optimizing the model according to the evaluation result.
S205, obtaining the expected residual life of each storage unit on the industrial solid state disk;
The management system may obtain the expected remaining life of each storage unit by reading life information on the controller firmware of the industrial grade solid state disk. Generally, when an industrial solid state disk leaves a factory, the controller firmware of the industrial solid state disk records the upper limit of the initial erasing times of each storage unit, and counts the used erasing times of each storage unit in real time, so as to evaluate and calculate the expected remaining life.
S206, if the predicted remaining life is lower than the preset life threshold, determining a storage unit with the predicted remaining life lower than the preset life threshold as a bad block, and isolating the bad block from the industrial solid state disk;
If the predicted remaining life is lower than the preset life threshold, the management system determines the storage unit with the predicted remaining life lower than the preset life threshold as a bad block and isolates the bad block from the industrial solid state disk. For example, the preset lifetime threshold is 80% of the total erasing times of the storage units, and when the predicted remaining lifetime of a certain storage unit indicates that the erasing times of the certain storage unit exceed 80% of the total erasing times, the management system directly determines that the storage unit is a bad block to isolate the storage unit, and other tests are not needed.
S207, if the predicted remaining life is higher than a preset life threshold, acquiring the surface temperature of each storage unit on the industrial solid state disk at a plurality of preset time points;
And if the predicted remaining life is higher than the preset life threshold, the management system acquires the surface temperature of each storage unit on the industrial solid state disk at a plurality of preset time points. In the example of the receiving step S202, for the storage unit with the expected remaining lifetime being greater than 80%, the management system may perform surface temperature acquisition at 1 to 2 points per day at a plurality of preset time points every ten minutes, and the detailed example will refer to step S101 and will not be described herein. The preset time point may be configured according to actual needs, and is not limited herein.
S208, arranging the surface temperatures of the storage units according to a time sequence to obtain surface temperature sets corresponding to the storage units respectively;
The detailed example of the surface temperature set corresponding to each storage unit may refer to step S101, which is not described herein.
S209, determining a storage unit meeting a preset abnormality judgment rule as a storage unit to be detected, wherein the preset abnormality judgment rule is that the values of the surface temperatures exceeding the preset number in the surface temperature set are not within a preset working temperature threshold;
specifically, refer to step S102, which is not described herein.
S210, acquiring the historical writing times, the historical writing intensity, the historical error rate and the historical health index of the training storage unit in the historical preset time;
The management system may extract usage data of each storage unit in a certain period (for example, a past preset period) from the usage log of the connected industrial solid state disk, where the usage data includes information such as the number of writing times, the writing strength (writing amount in unit time), the error rate, and the health index obtained by past detection, which are used as the historical number of writing times, the historical writing strength, the historical error rate, and the historical health index of the training storage unit in the historical preset period, and this step may refer to step S103, which is not described herein.
S211, marking the historical writing times, the historical writing strength and the historical error rate as input features, and marking the historical health index as output features;
the input features and the corresponding output features are used as a group of training data, so that training of the preset model is facilitated. The input features and the corresponding output features are used as a set of verification data, so that the preset model can be verified conveniently, and the step can refer to the step S103, which is not repeated here.
S212, training a preset model by using the input features and the output features to obtain the accuracy of the preset model;
The management system trains the preset model by using the input characteristics and the output characteristics to obtain the accuracy of the preset model. The predetermined model may be a deep neural network health assessment model. The training process continuously adjusts model parameters, and fits the mapping relation between the input characteristics and the output characteristics, so that the predicted output of the preset model can be close to the output of the real health index. The accuracy of the preset model is evaluated by the verification data, and this step may refer to step S103, which is not described herein.
S213, when the accuracy exceeds a preset accuracy threshold, a health assessment model is obtained;
And when the accuracy of the preset model exceeds a preset accuracy threshold, the management system obtains a health assessment model. For example, the preset accuracy threshold is 80%, and when the evaluation accuracy of the preset model in the verification data reaches 80%, the preset model can be used as a health evaluation model of the industrial solid state disk. The accuracy threshold may be set according to practical situations, and is not limited herein. This step may refer to step S103, and will not be described here again.
S214, inputting the writing times, the writing strength and the error rate of the storage unit to be detected in a preset time period into a preset health evaluation model to obtain the health index of the storage unit to be detected;
specifically, reference may be made to step S103, which is not described herein.
S215, when the health index is lower than a preset first threshold value and higher than a preset second threshold value, determining the storage unit to be detected as an early warning block, backing up the effective data in the early warning block, and executing data migration operation, wherein the data migration operation is used for migrating the effective data from the early warning block to other storage units;
specifically, reference may be made to step S104, which is not described herein.
S216, when the health index is lower than a preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block, and isolating the bad block from the industrial solid state disk, wherein the preset first threshold is larger than the preset second threshold;
Specifically, reference may be made to step S105, which is not described herein.
S217, determining the data detection frequency and the data migration rate of the early warning block according to the security levels of the stored data respectively corresponding to the storage units;
And the management system determines the data detection frequency and the data migration rate of the early warning block according to the security levels of the stored data respectively corresponding to the storage units. For the stored data with higher security level, the management system can set higher data detection frequency, such as detection once every 5 minutes, so as to monitor the storage environment of the early warning block in time. Meanwhile, the management system can set a faster data migration rate for an early warning block where the stored data with a higher security level is located, for example, the data migration is completed within 1 hour after the early warning state is detected, and the data migration can be completed within 24 hours for the stored data with a lower security level. The data detection frequency configuration and the data migration rate configuration based on the security level can ensure timely monitoring and quick remote backup of important data.
S218, determining isolation priority of bad blocks according to security levels of storage data corresponding to the storage units respectively;
And the management system determines the isolation priority of the bad blocks according to the security levels of the storage data respectively corresponding to the storage units. For the storage data with higher security level, when the storage unit is determined to be a bad block, the management system gives higher isolation priority and performs isolation processing as soon as possible. Whereas for stored data with a lower security level, the quarantine process may be performed with a lower priority. Meanwhile, if the number of bad blocks exceeds the capacity of the industrial solid state disk, the management system can preferentially isolate the bad blocks where the storage data with lower security level are located. Such isolation priority based on data security level configuration may ensure that the storage environment of important data is preferentially protected.
The following describes a management system in the embodiment of the present application from the perspective of hardware processing, please refer to fig. 3, which is a schematic diagram of a physical device structure of the management system in the embodiment of the present application.
It should be noted that the structure of the management system shown in fig. 3 is only an example, and should not impose any limitation on the functions and the application scope of the embodiments of the present invention.
As shown in fig. 3, the management system includes a central processing unit (Central Processing Unit, CPU) 301 that can perform various appropriate actions and processes, such as performing the methods described in the above embodiments, according to a program stored in a Read-Only Memory (ROM) 302 or a program loaded from a storage section 308 into a random access Memory (Random Access Memory, RAM) 303. In the RAM 303, various programs and data required for the system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other through a bus 304. An Input/Output (I/O) interface 305 is also connected to bus 304.
The following components are connected to the I/O interface 305: an input section 306 including an audio input device, a push button switch, and the like; an output portion 307 including a Liquid crystal display (Liquid CRYSTAL DISPLAY, LCD), an audio output device, an indicator lamp, and the like; a storage section 308 including a hard disk or the like; and a communication section 309 including a network interface card such as a LAN (Local Area Network ) card, a modem, or the like. The communication section 309 performs communication processing via a network such as the internet. The drive 310 is also connected to the I/O interface 305 as needed. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed on the drive 310 as needed, so that a computer program read therefrom is installed into the storage section 308 as needed.
In particular, according to embodiments of the present invention, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising a computer program for performing the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication portion 309, and/or installed from the removable medium 311. When the computer program is executed by a Central Processing Unit (CPU) 301, various functions defined in the present invention are performed.
Specific examples of the computer-readable storage medium include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-Only Memory (ROM), an erasable programmable read-Only Memory (Erasable Programmable Read Only Memory, EPROM), a flash Memory, an optical fiber, a portable compact disc read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. Where each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures.
Specifically, the management system of the embodiment includes a processor and a memory, where the memory stores a computer program, and when the computer program is executed by the processor, the bad block management method of the industrial solid state disk provided in the embodiment is implemented.
As another aspect, the present invention also provides a computer-readable storage medium, which may be contained in the management system described in the above embodiment; or may exist alone without being assembled into the management system. The storage medium carries one or more computer programs which, when executed by a processor of the management system, cause the management system to implement the bad block management method of the industrial solid state disk provided in the above embodiment.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application.
As used in the above embodiments, the term "when …" may be interpreted to mean "if …" or "after …" or "in response to determination …" or "in response to detection …" depending on the context. Similarly, the phrase "at the time of determination …" or "if detected (a stated condition or event)" may be interpreted to mean "if determined …" or "in response to determination …" or "at the time of detection (a stated condition or event)" or "in response to detection (a stated condition or event)" depending on the context.
Those of ordinary skill in the art will appreciate that implementing all or part of the above-described method embodiments may be accomplished by a computer program to instruct related hardware, the program may be stored in a computer readable storage medium, and the program may include the above-described method embodiments when executed. And the aforementioned storage medium includes: ROM or random access memory RAM, magnetic or optical disk, etc.
Claims (10)
1. The bad block management method of the industrial solid state disk is characterized by being applied to a management system, and comprises the following steps:
Periodically acquiring the surface temperature of each storage unit on the industrial solid state disk to obtain a surface temperature set corresponding to each storage unit;
determining a storage unit meeting a preset abnormality judgment rule as a storage unit to be detected, wherein the preset abnormality judgment rule is that the values of the surface temperatures exceeding the preset number in the surface temperature set are not within a preset working temperature threshold;
Inputting the writing times, the writing intensity and the error rate of the storage unit to be detected in a preset time period into a preset health evaluation model to obtain the health index of the storage unit to be detected;
When the health index is lower than a preset first threshold value and higher than a preset second threshold value, determining the storage unit to be detected as an early warning block, backing up effective data in the early warning block and executing data migration operation, wherein the data migration operation is used for migrating the effective data from the early warning block to other storage units;
And when the health index is lower than the preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block and isolating the bad block from the industrial solid state disk, wherein the preset first threshold is larger than the preset second threshold.
2. The method according to claim 1, wherein when the health index is lower than the preset second threshold, determining the storage unit to be detected as a bad block, stopping the read-write activity of the bad block, and isolating the bad block from the industrial solid state disk, specifically includes:
when the health index is lower than the preset second threshold value, determining the storage unit to be detected as a bad block;
searching a logical address corresponding to the physical address of the bad block according to the physical address of the bad block;
Marking the physical address of the bad block as forbidden, and completing the operation of isolating the bad block from the industrial solid state disk;
Marking the data corresponding to the logical address as damaged data to stop the read-write activity of the bad block.
3. The method of claim 1, wherein before the step of periodically obtaining the surface temperatures of the storage units on the industrial solid state disk to obtain the surface temperature sets respectively corresponding to the storage units, the method further comprises:
obtaining the predicted residual life of each storage unit on the industrial solid state disk;
And if the predicted remaining life is lower than a preset life threshold, determining the storage unit with the predicted remaining life lower than the preset life threshold as a bad block, and isolating the bad block from the industrial solid state disk.
4. The method of claim 3, wherein the periodically obtaining the surface temperature of each storage unit on the industrial solid state disk to obtain the surface temperature set corresponding to each storage unit respectively specifically includes:
if the predicted remaining life is higher than a preset life threshold, acquiring the surface temperature of each storage unit on the industrial solid state disk at a plurality of preset time points;
and arranging the surface temperatures of the storage units according to a time sequence to obtain the surface temperature sets corresponding to the storage units respectively.
5. The method of claim 1, wherein before the step of periodically obtaining the surface temperatures of the storage units on the industrial solid state disk to obtain the surface temperature sets respectively corresponding to the storage units, the method further comprises:
Acquiring storage data respectively corresponding to each storage unit stored on the industrial solid state disk;
Extracting a plurality of characteristic information of the stored data, wherein the characteristic information comprises a data format, a data size, a data creation time, a data naming mode and data creator information;
performing matching analysis on the plurality of characteristic information based on a preset information base, and determining the security level of the stored data;
if the matching analysis fails, the plurality of characteristic information is input into a preset data evaluation model to obtain a security level score and a corresponding security level.
6. The method of claim 5, wherein the method further comprises:
Determining the data detection frequency and the data migration rate of the early warning block according to the security levels of the stored data respectively corresponding to the storage units;
and determining the isolation priority of the bad blocks according to the security levels of the stored data respectively corresponding to the storage units.
7. The method according to claim 1, wherein before the step of inputting the number of writing times, the writing strength and the error rate of the storage unit to be detected in the preset duration into a preset health assessment model to obtain the health index of the storage unit to be detected, the method further comprises:
acquiring the historical writing times, the historical writing intensity, the historical error rate and the historical health index of the training storage unit in the historical preset time;
marking the historical writing times, the historical writing intensity and the historical error rate as input features, and marking the historical health index as output features;
training a preset model by using the input features and the output features to obtain the accuracy of the preset model;
and when the accuracy exceeds a preset accuracy threshold, obtaining a health assessment model.
8. A management system, the management system comprising: one or more processors and memory; the memory is coupled to the one or more processors, the memory for storing computer program code comprising computer instructions that the one or more processors invoke to cause the management system to perform the method of any of claims 1-7.
9. A computer readable storage medium comprising instructions which, when run on a management system, cause the management system to perform the method of any of claims 1-7.
10. A computer program product, characterized in that the computer program product, when run on a management system, causes the management system to perform the method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410734289.3A CN118312109B (en) | 2024-06-07 | 2024-06-07 | Bad block management method, system, medium and product of industrial solid state disk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410734289.3A CN118312109B (en) | 2024-06-07 | 2024-06-07 | Bad block management method, system, medium and product of industrial solid state disk |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118312109A true CN118312109A (en) | 2024-07-09 |
| CN118312109B CN118312109B (en) | 2024-09-03 |
Family
ID=91724880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410734289.3A Active CN118312109B (en) | 2024-06-07 | 2024-06-07 | Bad block management method, system, medium and product of industrial solid state disk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118312109B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106294065A (en) * | 2016-07-28 | 2017-01-04 | 联想(北京)有限公司 | Hard disk failure monitoring method, Apparatus and system |
| CN109408328A (en) * | 2018-10-08 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of monitoring method, device and the equipment of hard disk health status |
| CN116501523A (en) * | 2023-03-24 | 2023-07-28 | 苏州浪潮智能科技有限公司 | Abnormality processing method and device for equipment, storage medium and electronic device |
| CN116631487A (en) * | 2023-05-26 | 2023-08-22 | 北京市合芯数字科技有限公司 | Solid state disk life prediction method, device, terminal and medium based on long-short-term memory network model |
| CN117573444A (en) * | 2023-10-30 | 2024-02-20 | 苏州元脑智能科技有限公司 | Data processing method, device, equipment and medium of storage system |
| CN118051948A (en) * | 2024-04-16 | 2024-05-17 | 深圳迅策科技股份有限公司 | Dynamic safe storage method based on big data platform |
| US20240168835A1 (en) * | 2022-11-23 | 2024-05-23 | Army Academy of Armored Forces | Hard disk failure prediction method, system, device and medium |
| CN118093324A (en) * | 2024-04-28 | 2024-05-28 | 深圳市领德创科技有限公司 | Solid state disk life prediction method and system |
-
2024
- 2024-06-07 CN CN202410734289.3A patent/CN118312109B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106294065A (en) * | 2016-07-28 | 2017-01-04 | 联想(北京)有限公司 | Hard disk failure monitoring method, Apparatus and system |
| CN109408328A (en) * | 2018-10-08 | 2019-03-01 | 郑州云海信息技术有限公司 | A kind of monitoring method, device and the equipment of hard disk health status |
| US20240168835A1 (en) * | 2022-11-23 | 2024-05-23 | Army Academy of Armored Forces | Hard disk failure prediction method, system, device and medium |
| CN116501523A (en) * | 2023-03-24 | 2023-07-28 | 苏州浪潮智能科技有限公司 | Abnormality processing method and device for equipment, storage medium and electronic device |
| CN116631487A (en) * | 2023-05-26 | 2023-08-22 | 北京市合芯数字科技有限公司 | Solid state disk life prediction method, device, terminal and medium based on long-short-term memory network model |
| CN117573444A (en) * | 2023-10-30 | 2024-02-20 | 苏州元脑智能科技有限公司 | Data processing method, device, equipment and medium of storage system |
| CN118051948A (en) * | 2024-04-16 | 2024-05-17 | 深圳迅策科技股份有限公司 | Dynamic safe storage method based on big data platform |
| CN118093324A (en) * | 2024-04-28 | 2024-05-28 | 深圳市领德创科技有限公司 | Solid state disk life prediction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118312109B (en) | 2024-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Which log level should developers choose for a new logging statement? | |
| Li et al. | Hard drive failure prediction using decision trees | |
| Li et al. | Towards just-in-time suggestions for log changes | |
| KR101948634B1 (en) | Failure prediction method of system resource for smart computing | |
| Shang et al. | Automated detection of performance regressions using regression models on clustered performance counters | |
| US20150121136A1 (en) | System and method for automatically managing fault events of data center | |
| CN110633567B (en) | RASP-based intelligent contract virtual machine vulnerability detection system and method | |
| Wang et al. | Data-driven risk assessment on urban pipeline network based on a cluster model | |
| CN117742618B (en) | Data storage management method and device for solid state disk | |
| CN117094184B (en) | Modeling method, system and medium of risk prediction model based on intranet platform | |
| CN111813585A (en) | Prediction and processing of slow discs | |
| WO2021094852A1 (en) | Alerting to model degradation based on distribution analysis using risk tolerance ratings | |
| CN113760670A (en) | Cable joint abnormity early warning method and device, electronic equipment and storage medium | |
| CN116389235A (en) | Fault monitoring method and system applied to industrial Internet of things | |
| CN115952081A (en) | Software testing method, device, storage medium and equipment | |
| CN118312109B (en) | Bad block management method, system, medium and product of industrial solid state disk | |
| CN114547640A (en) | Method and device for judging sensitive operation behaviors, electronic equipment and storage medium | |
| KR20220118621A (en) | Machine Learning based Security Management System for Collaborative Robot in the Machine Zone | |
| CN114091644A (en) | Technical risk assessment method and system for artificial intelligence product | |
| CN110865939A (en) | Application program quality monitoring method and device, computer equipment and storage medium | |
| Rombach et al. | SmartPred: Unsupervised hard disk failure detection | |
| JP2019133470A (en) | Behavior discrimination method, behavior discrimination device and program | |
| Zhu et al. | A survey on default prediction of cloud storage service | |
| US11810013B2 (en) | Systems and methods for alerting to model degradation based on survival analysis | |
| US11256597B2 (en) | Ensemble approach to alerting to model degradation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |