CN109857607A - A kind of reliability checking method and device of NAND Flash solid state hard disk - Google Patents
A kind of reliability checking method and device of NAND Flash solid state hard disk Download PDFInfo
- Publication number
- CN109857607A CN109857607A CN201811583676.2A CN201811583676A CN109857607A CN 109857607 A CN109857607 A CN 109857607A CN 201811583676 A CN201811583676 A CN 201811583676A CN 109857607 A CN109857607 A CN 109857607A
- Authority
- CN
- China
- Prior art keywords
- bitmap
- detected
- hard disk
- solid state
- state hard
- 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.)
- Pending
Links
Landscapes
- For Increasing The Reliability Of Semiconductor Memories (AREA)
Abstract
The present invention discloses the reliability checking method and device of a kind of NAND Flash solid state hard disk, this method comprises: obtaining the corresponding bitmap of solid state hard disk to be detected；Neural network model trained according to the bitmap and in advance carries out reliability detection to the solid state hard disk to be detected.The present invention combines bitmap analysis with neural network, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance, the memory block that bitmap and solid state hard disk to be detected include corresponds, it realizes and quick fail-safe analysis is carried out to each memory block that solid state hard disk includes, substantially increase the speed and efficiency of fail-safe analysis, the test period for shortening solid state hard disk, production life cycle is shortened, has saved cost.
Description
Technical field
The invention belongs to technical field of integrated circuits, and in particular to a kind of reliability detection of NAND Flash solid state hard disk
Method and device.
Background technique
SSD (Solid State Disk, solid state hard disk) technology based on NAND Flash (nand flash memory) is answered extensively
Used in fields such as automotive electronics, family life, Industry Control, enterprise's application and smart homes.It is needed pair before SSD factory
SSD carries out reliability detection, and the reliability of the SSD to ensure to dispatch from the factory is up to standard.
Currently, a kind of method for detecting the SSD reliability based on NAND Flash is proposed in the related technology, and this method makes
Fail-safe analysis is carried out to NAND Flash with Bitmap (bitmap), discusses peripheral circuit, array organization's form and technique
Influence to NAND Flash reliability.
But quickly may not be used on a large scale to Bitmap progress when above-mentioned the relevant technologies carry out fail-safe analysis using Bitmap
It is analyzed by property, inefficiency, in the case where current huge storage density increases, the period is too long, and cost is too high.
Summary of the invention
In order to solve the above problem, the present invention provides the reliability checking method and dress of a kind of NAND Flash solid state hard disk
It sets, bitmap analysis is combined with neural network, solid-state to be detected is detected by neural network model trained in advance automatically
The reliability of the corresponding bitmap of hard disk is realized and carries out quick fail-safe analysis to each memory block that solid state hard disk includes.This
Invention solves problem above by the following aspects.
In a first aspect, the embodiment of the invention provides a kind of reliability checking method of NAND Flash solid state hard disk, institute
The method of stating includes:
Obtain the corresponding bitmap of solid state hard disk to be detected；
Neural network model trained according to the bitmap and in advance carries out reliability inspection to the solid state hard disk to be detected
It surveys.
With reference to first aspect, the embodiment of the invention provides the first possible implementation of above-mentioned first aspect,
In, the neural network model trained according to the bitmap and in advance carries out reliability inspection to the solid state hard disk to be detected
It surveys, comprising:
The neural network model that the corresponding each bitmap input of the solid state hard disk to be detected is trained in advance respectively, obtains
The reliability testing result of the corresponding memory block of each bitmap；
According to the reliability testing result of each memory block, it is default to determine whether the solid state hard disk to be detected meets
Reliable condition.
With reference to first aspect, the embodiment of the invention provides second of possible implementation of above-mentioned first aspect,
In, it is described to obtain the corresponding bitmap of solid state hard disk to be detected, comprising:
Determine each memory block that solid state hard disk to be detected includes；
Each memory block is scanned respectively, obtains the corresponding bitmap of each memory block.
With reference to first aspect, the embodiment of the invention provides the third possible implementation of above-mentioned first aspect,
In, the neural network model trained according to the bitmap and in advance carries out reliability inspection to the solid state hard disk to be detected
Before survey, further includes:
It obtains the bitmap of detection of preset quantity and each described has detected the corresponding reliability testing result of bitmap；
By the bitmap of detection of the preset quantity and each described the corresponding reliability testing result of bitmap is detected
Training neural network model.
The third possible implementation with reference to first aspect, the embodiment of the invention provides the of above-mentioned first aspect
Four kinds of possible implementations, wherein the bitmap of detection by the preset quantity and each described detected bitmap
Corresponding reliability testing result training neural network model, comprising:
It has detected threshold voltage information or mutual conductance information that bitmap includes by each and each described has detected bitmap pair
The reliability testing result training neural network model answered.
The possible implementation of with reference to first aspect the first, the embodiment of the invention provides the of above-mentioned first aspect
Five kinds of possible implementations, wherein the reliability testing result according to each memory block determines described to be detected
Whether solid state hard disk, which meets, is preset reliable condition, comprising:
According to the reliability testing result of each memory block, the corresponding memory block of the solid state hard disk to be detected is calculated
Reliable sex rate；
If the reliable sex rate of memory block is greater than preset threshold, it is determined that the solid state hard disk satisfaction to be detected is default can
By condition.
Second aspect, the embodiment of the invention provides a kind of reliability detecting device of NAND Flash solid state hard disk, institutes
Stating device includes:
Module is obtained, for obtaining the corresponding bitmap of solid state hard disk to be detected；
Reliability detection module, for neural network model trained according to the bitmap and in advance, to described to be detected
Solid state hard disk carries out reliability detection.
In conjunction with second aspect, the embodiment of the invention provides the first possible implementation of above-mentioned second aspect,
In, the reliability detection module includes:
Input unit, for respectively that the corresponding each bitmap input of the solid state hard disk to be detected is trained in advance nerve
Network model obtains the reliability testing result of the corresponding memory block of each bitmap；
Determination unit determines that the solid-state to be detected is hard for the reliability testing result according to each memory block
Whether disk, which meets, is preset reliable condition.
In conjunction with second aspect, the embodiment of the invention provides second of possible implementation of above-mentioned second aspect,
In, described device further include:
Model training module, for obtain preset quantity the bitmap of detection and it is each described detected that bitmap is corresponding can
By property testing result；It is detected by the bitmap of detection and each corresponding reliability of bitmap that detected of the preset quantity
As a result neural network model is trained.
The third aspect, the embodiment of the invention provides a kind of reliability detection device of NAND Flash solid state hard disk, packets
It includes:
One or more processors；
Storage device, for storing one or more programs；
One or more of programs are executed by one or more of processors, so that one or more of processors
Realize method described in any possible implementation of above-mentioned first aspect or first aspect.
In embodiments of the present invention, the corresponding bitmap of solid state hard disk to be detected is obtained；According to the bitmap and training in advance
Neural network model, reliability detection is carried out to the solid state hard disk to be detected.The present invention is by bitmap analysis and neural network
Combine, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance,
The memory block that bitmap and solid state hard disk to be detected include corresponds, and realizes and carries out fastly to each memory block that solid state hard disk includes
The fail-safe analysis of speed, substantially increases the speed and efficiency of fail-safe analysis, shortens the test period of solid state hard disk, shortens
Production life cycle, has saved cost.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows a kind of reliability checking method of NAND Flash solid state hard disk provided by the embodiment of the present invention 1
Flow diagram；
Fig. 2 shows the schematic diagrames of the training of CNN network model provided by the embodiment of the present invention 1；
Fig. 3 shows the reliability detection side of another kind NAND Flash solid state hard disk provided by the embodiment of the present invention 1
The flow diagram of method.
Fig. 4 shows a kind of reliability detecting device of NAND Flash solid state hard disk provided by the embodiment of the present invention 2
Structural schematic diagram.
Specific embodiment
The illustrative embodiments of the disclosure are more fully described below with reference to accompanying drawings.Although showing this public affairs in attached drawing
The illustrative embodiments opened, it being understood, however, that may be realized in various forms the disclosure without the reality that should be illustrated here
The mode of applying is limited.It is to be able to thoroughly understand the disclosure on the contrary, providing these embodiments, and can be by this public affairs
The range opened is fully disclosed to those skilled in the art.
Embodiment 1
The embodiment of the invention provides a kind of reliability checking method of NAND Flash solid state hard disk, this method is by bitmap
Analysis combines with neural network, by neural network model trained in advance to the corresponding position of NAND Flash solid state hard disk
Figure carries out automatic fail-safe analysis, realizes each array for including to NAND Flash solid state hard disk from product level and does fastly
Fast fail-safe analysis substantially increases the speed of fail-safe analysis, shortens test period, improves detection efficiency, and reduce
Testing cost.
One NAND Flash solid state hard disk includes multiple memory blocks (Block), and each memory block respectively corresponds a position
Figure, the corresponding bitmap of NAND Flash solid state hard disk is the corresponding position of each memory block that NAND Flash solid state hard disk includes
Figure.Before practical application this method carries out the detection of solid state hard disk reliability, as shown in Figure 1, being instructed first by operating as follows
Practice neural network model, specifically include:
A1: obtaining the bitmap of detection of preset quantity and has each detected the corresponding reliability testing result of bitmap.A2: logical
It crosses the bitmap of detection of preset quantity and has each detected the corresponding reliability testing result training neural network model of bitmap.
The accuracy of the neural network model of the bigger training of the value of above-mentioned preset quantity is higher, and the value of preset quantity is got over
It is small, then train the speed of neural network model faster.It can determine the value of the preset quantity according to demand in practical applications.
Having detected of above-mentioned preset quantity needed in bitmap include reliability testing result be true or 1 the inspection of bitmap and reliability
Surveying result is false or 0 bitmap.That is it needs to include simultaneously reliable bitmap and insecure bitmap, and the two need to respectively account for one
Certainty ratio.For example, it is assumed that preset quantity is 100, then reliable bitmap is 50, and insecure bitmap is 50；Alternatively, reliable position
Figure is 60, and insecure bitmap is 40 etc..
Each having detected the corresponding reliability testing result of bitmap is by artificial detection, based on Data Retention
What the detections such as the test mode of (data preservation) or the test mode for being based on Endurance (durability) obtained.
By in the bitmap of the detection input neural network model of preset quantity, neural network model has detected bitmap to these
It is trained study, output has each detected the corresponding model test results of bitmap.Bitmap has been detected for each, has compared this
It is whether consistent with the reliability testing result of above-mentioned acquisition to detect the corresponding model test results of bitmap, it is consistent to calculate two results
The bitmap that detected detected in bitmap shared ratio all, which can embody neural network model and carry out reliability
The accuracy of detection.When the ratio is greater than default value, show that the detection accuracy of Current Situation of Neural Network model meets the requirements,
Stop the training to neural network model.
In embodiments of the present invention, neural network model can use CNN (Convolutional Neural
Networks, convolutional neural networks), BPNN (Back Propagation Neural Networks, BP neural network predict mould
Type) or other neural network forms.
When neural network uses CNN, CNN includes convolutional layer, pond layer, input layer, hidden layer and output layer.Wherein,
Multiple convolutional layers and pond layer can be set according to demand before input layer, if CNN successively includes convolutional layer, pond layer, convolution
Layer, pond layer, input layer, hidden layer and output layer.Multiple groups convolutional layer and pond layer so arranged in a crossed manner plays crossing filtering
The effect of data.Hidden layer includes convolutional layer, pond layer and full articulamentum, and hidden layer is arranged according to demand in practical application includes
Which layer, the connection relationship of each layer and number of plies of each layer etc..The number of plies of convolutional layer, the number of plies of pond layer, the number of convolution kernel
It can be arranged according to demand with the interstitial content of size, the number of plies of hidden layer, input layer and output layer etc., convolutional layer and pond
The number of plies for changing layer setting is more, and the feature grabbed from the bitmap of input is more.
In CNN network model training schematic diagram as shown in Figure 2, bitmap input CNN net has been detected to N number of in will acquire 1
In network, which successively includes the first convolutional layer, the first pond layer, the second convolutional layer, the second pond layer, input layer, implies
Layer and output layer.Wherein, the first convolutional layer and the first pond layer include three layers, and the second convolutional layer and the second pond layer include
Six layers, input layer includes 8 input nodes, and output layer includes 2 output nodes.It can be according to the size of memory block in practical application
The picture formats such as length-width ratio for changing bitmap piece, adjust according to demand the convolution number of plies and the pond number of plies and convolution kernel number and
Size, and the implicit number of plies is adjusted according to demand.
After being trained by the bitmap of detection of the CNN network model shown in Fig. 2 to acquisition, it can obtain such as table 1
Shown in store bitmap block and model test results the table of comparisons.It is subsequent for each storage bitmap block, it is corresponding to compare its
Whether model test results are consistent with the reliability testing result of above-mentioned acquisition, calculate the consistent bitmap that detected of two results and exist
It is all to have detected ratio shared in bitmap, when the ratio is greater than default value, stop the training to neural network model.
Table 1
| Store bitmap block | Model test results |
| Block[1] | 1 (reliable) |
| Block[2] | 0 (unreliable) |
| ... | ... |
| Block[N-1] | 0 (unreliable) |
| Block[N] | 1 (reliable) |
In embodiments of the present invention, each of bitmap pixel represents the threshold voltage information of a storage unit.
It can detect bitmap includes threshold voltage information by each at training neural network model and each detected bitmap
Corresponding reliability testing result trains neural network model.Each mutual conductance information for having detected bitmap and having included can also be passed through
And the corresponding reliability testing result of bitmap each has been detected to train neural network model.
After training the neural network model for carrying out solid state hard disk reliability detection through the above way, referring to Fig. 3,
101 and 102 operation to carry out reliability detection to solid state hard disk to be detected as follows:
Step 101: obtaining the corresponding bitmap of solid state hard disk to be detected.
Determine each memory block that solid state hard disk to be detected includes；Each memory block is scanned respectively, is obtained each
The corresponding bitmap of memory block.
Step 102: neural network model trained according to the bitmap of acquisition and in advance, it can to solid state hard disk to be detected progress
By property detection.
Reliability detection is carried out to solid state hard disk to be detected especially by the operation of following steps S1 and S2, comprising:
S1: the neural network model that the corresponding each bitmap input of solid state hard disk to be detected is trained in advance respectively obtains
The reliability testing result of the corresponding memory block of each bitmap.
In practical applications, the neural network mould that the corresponding each bitmap input of solid state hard disk to be detected is trained in advance
Type, neural network model detect each bitmap of input, export the corresponding reliability testing result of each bitmap.Bitmap
Corresponding reliability testing result just represents the reliability of the corresponding memory block of the bitmap.
The embodiment of the present invention only need to pass through artificial detection in the training neural network model stage, be based on Data
The test mode of Retention or the test mode based on Endurance etc. detect the bitmap reliability inspection of preset quantity
It surveys as a result, later reliability detection can be carried out automatically by trained neural network model in practical applications, greatly improves
The speed and efficiency of reliability detection, shorten test period, have saved cost.
S2: according to the reliability testing result of each memory block, it is default reliable to determine whether solid state hard disk to be detected meets
Condition.
According to the reliability testing result of each memory block, the corresponding memory block reliability ratio of solid state hard disk to be detected is calculated
Rate.The corresponding reliable sex rate of memory block of solid state hard disk to be detected be reliability testing result be true or 1 memory block number
Ratio between the total number for the memory block for including with solid state hard disk to be detected.If the reliable sex rate of memory block is greater than default threshold
Value, it is determined that solid state hard disk satisfaction to be detected presets reliable condition, then the solid state hard disk to be detected is detected by reliability.If depositing
It stores up the reliable sex rate of block and is less than or equal to preset threshold, it is determined that solid state hard disk to be detected is unsatisfactory for presetting reliable condition, then should
Solid state hard disk to be detected is not detected by reliability.Above-mentioned preset threshold can be 70%, 80% or 90% etc..
In embodiments of the present invention, the corresponding bitmap of solid state hard disk to be detected is obtained；According to the bitmap and training in advance
Neural network model, reliability detection is carried out to the solid state hard disk to be detected.The present invention is by bitmap analysis and neural network
Combine, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance,
The memory block that bitmap and solid state hard disk to be detected include corresponds, and realizes and carries out fastly to each memory block that solid state hard disk includes
The fail-safe analysis of speed, substantially increases the speed and efficiency of fail-safe analysis, shortens the test period of solid state hard disk, shortens
Production life cycle, has saved cost.
Embodiment 2
Referring to fig. 4, the embodiment of the invention provides a kind of reliability detecting device of NAND Flash solid state hard disk, the dresses
The reliability checking method for executing NAND Flash solid state hard disk provided by above-described embodiment 1 is set, which includes:
Module 20 is obtained, for obtaining the corresponding bitmap of solid state hard disk to be detected；
Reliability detection module 21 is hard to solid-state to be detected for neural network model trained according to bitmap and in advance
Disk carries out reliability detection.
Above-mentioned reliability detection module 21 includes:
Input unit, for respectively that the corresponding each bitmap input of solid state hard disk to be detected is trained in advance neural network
Model obtains the reliability testing result of the corresponding memory block of each bitmap；
Determination unit determines whether solid state hard disk to be detected is full for the reliability testing result according to each memory block
Foot presets reliable condition.
Above-mentioned determination unit calculates solid state hard disk pair to be detected for the reliability testing result according to each memory block
The reliable sex rate of the memory block answered；If the reliable sex rate of memory block is greater than preset threshold, it is determined that solid state hard disk to be detected meets
Preset reliable condition.
Above-mentioned acquisition module 20 includes:
Memory block determination unit, each memory block for including for determining solid state hard disk to be detected；
Scanning element obtains the corresponding bitmap of each memory block for being scanned respectively to each memory block.
In embodiments of the present invention, the device further include: model training module, for obtaining the check bit of preset quantity
Scheme and each detected the corresponding reliability testing result of bitmap；Pass through the bitmap of detection and each check bit of preset quantity
Scheme corresponding reliability testing result training neural network model.
Above-mentioned model training module, for by each detected threshold voltage information or mutual conductance information that bitmap includes and
The corresponding reliability testing result training neural network model of bitmap is each detected.
In embodiments of the present invention, the corresponding bitmap of solid state hard disk to be detected is obtained；According to the bitmap and training in advance
Neural network model, reliability detection is carried out to the solid state hard disk to be detected.The present invention is by bitmap analysis and neural network
Combine, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance,
The memory block that bitmap and solid state hard disk to be detected include corresponds, and realizes and carries out fastly to each memory block that solid state hard disk includes
The fail-safe analysis of speed, substantially increases the speed and efficiency of fail-safe analysis, shortens the test period of solid state hard disk, shortens
Production life cycle, has saved cost.
Embodiment 3
The embodiment of the present invention provides a kind of reliability detection device of NAND Flash solid state hard disk, which includes one
Or multiple processors, and one or more storage device are stored with one or more in one or more of storage devices
Program when one or more of programs are loaded and executed by one or more of processors, realizes that above-described embodiment 1 is mentioned
The reliability checking method of the NAND Flash solid state hard disk of confession.
In embodiments of the present invention, the corresponding bitmap of solid state hard disk to be detected is obtained；According to the bitmap and training in advance
Neural network model, reliability detection is carried out to the solid state hard disk to be detected.The present invention is by bitmap analysis and neural network
Combine, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance,
The memory block that bitmap and solid state hard disk to be detected include corresponds, and realizes and carries out fastly to each memory block that solid state hard disk includes
The fail-safe analysis of speed, substantially increases the speed and efficiency of fail-safe analysis, shortens the test period of solid state hard disk, shortens
Production life cycle, has saved cost.
Embodiment 4
The embodiment of the present invention provide a kind of computer can storage medium, be stored with executable program in the storage medium, institute
State executable code processor load and realize NAND Flash solid state hard disk provided by above-described embodiment 1 when executing can
By property detection method.
In embodiments of the present invention, the corresponding bitmap of solid state hard disk to be detected is obtained；According to the bitmap and training in advance
Neural network model, reliability detection is carried out to the solid state hard disk to be detected.The present invention is by bitmap analysis and neural network
Combine, detect the reliability of the corresponding bitmap of solid state hard disk to be detected automatically by neural network model trained in advance,
The memory block that bitmap and solid state hard disk to be detected include corresponds, and realizes and carries out fastly to each memory block that solid state hard disk includes
The fail-safe analysis of speed, substantially increases the speed and efficiency of fail-safe analysis, shortens the test period of solid state hard disk, shortens
Production life cycle, has saved cost.
It should be understood that
Algorithm and display do not have intrinsic phase with any certain computer, virtual bench or other equipment provided herein
It closes.Various fexible units can also be used together with teachings based herein.As described above, this kind of device is constructed to be wanted
The structure asked is obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use each
Kind programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this
The preferred forms of invention.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e. required to protect
Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
Meaning one of can in any combination mode come using.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
One in the creating device of microprocessor or digital signal processor (DSP) to realize virtual machine according to an embodiment of the present invention
The some or all functions of a little or whole components.The present invention is also implemented as executing method as described herein
Some or all device or device programs (for example, computer program and computer program product).Such realization
Program of the invention can store on a computer-readable medium, or may be in the form of one or more signals.This
The signal of sample can be downloaded from an internet website to obtain, and is perhaps provided on the carrier signal or mentions in any other forms
For.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of the claim
Subject to enclosing.
Claims (10)
1. a kind of reliability checking method of NAND Flash solid state hard disk, which is characterized in that the described method includes:
Obtain the corresponding bitmap of solid state hard disk to be detected；
Neural network model trained according to the bitmap and in advance carries out reliability detection to the solid state hard disk to be detected.
2. the method according to claim 1, wherein the neural network trained according to the bitmap and in advance
Model carries out reliability detection to the solid state hard disk to be detected, comprising:
The neural network model that the corresponding each bitmap input of the solid state hard disk to be detected is trained in advance respectively, obtains each
The reliability testing result of the corresponding memory block of the bitmap；
According to the reliability testing result of each memory block, it is default reliable to determine whether the solid state hard disk to be detected meets
Condition.
3. the method according to claim 1, wherein described obtain the corresponding bitmap of solid state hard disk to be detected, packet
It includes:
Determine each memory block that solid state hard disk to be detected includes；
Each memory block is scanned respectively, obtains the corresponding bitmap of each memory block.
4. method according to claim 1-3, which is characterized in that according to the bitmap and in advance training
Neural network model, before the solid state hard disk progress reliability detection to be detected, further includes:
It obtains the bitmap of detection of preset quantity and each described has detected the corresponding reliability testing result of bitmap；
It is trained by the bitmap of detection and each corresponding reliability testing result of bitmap that detected of the preset quantity
Neural network model.
5. according to the method described in claim 4, it is characterized in that, the bitmap of detection by the preset quantity and every
A corresponding reliability testing result of bitmap that detected trains neural network model, comprising:
Threshold voltage information or mutual conductance information that bitmap includes have been detected and each described to have detected bitmap corresponding by each
Reliability testing result trains neural network model.
6. according to the method described in claim 2, it is characterized in that, described detected according to the reliability of each memory block is tied
Fruit determines whether the solid state hard disk to be detected meets and presets reliable condition, comprising:
According to the reliability testing result of each memory block, it is reliable to calculate the corresponding memory block of the solid state hard disk to be detected
Sex rate；
If the reliable sex rate of memory block is greater than preset threshold, it is determined that the solid state hard disk satisfaction to be detected presets reliable item
Part.
7. a kind of reliability detecting device of NAND Flash solid state hard disk, which is characterized in that described device includes:
Module is obtained, for obtaining the corresponding bitmap of solid state hard disk to be detected；
Reliability detection module, for neural network model trained according to the bitmap and in advance, to the solid-state to be detected
Hard disk carries out reliability detection.
8. device according to claim 7, which is characterized in that the reliability detection module includes:
Input unit, for respectively that the corresponding each bitmap input of the solid state hard disk to be detected is trained in advance neural network
Model obtains the reliability testing result of the corresponding memory block of each bitmap；
Determination unit determines that the solid state hard disk to be detected is for the reliability testing result according to each memory block
No satisfaction presets reliable condition.
9. device according to claim 7 or 8, which is characterized in that described device further include:
Model training module, for obtaining the bitmap of detection of preset quantity and each described having detected the corresponding reliability of bitmap
Testing result；By the bitmap of detection of the preset quantity and each described the corresponding reliability testing result of bitmap is detected
Training neural network model.
10. a kind of reliability detection device of NAND Flash solid state hard disk characterized by comprising
One or more processors；
Storage device, for storing one or more programs；
One or more of programs are executed by one or more of processors, so that one or more of processors are realized
Such as method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583676.2A CN109857607A (en) | 2018-12-24 | 2018-12-24 | A kind of reliability checking method and device of NAND Flash solid state hard disk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583676.2A CN109857607A (en) | 2018-12-24 | 2018-12-24 | A kind of reliability checking method and device of NAND Flash solid state hard disk |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109857607A true CN109857607A (en) | 2019-06-07 |
Family
ID=66891989
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811583676.2A Pending CN109857607A (en) | 2018-12-24 | 2018-12-24 | A kind of reliability checking method and device of NAND Flash solid state hard disk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109857607A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580932A (en) * | 2019-08-29 | 2019-12-17 | 华中科技大学 | Memory cell quality measurement method applied to wear leveling |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964213A (en) * | 2009-07-24 | 2011-02-02 | 中芯国际集成电路制造(上海)有限公司 | Test method on failure analysis of storage cell |
| US20160104539A1 (en) * | 2014-10-08 | 2016-04-14 | Kyungryun Kim | Storage device and reliability verification method |
| CN107967928A (en) * | 2017-10-18 | 2018-04-27 | 武汉忆数存储技术有限公司 | A kind of flash chip life-span prediction method based on mathematical model |
| CN108647136A (en) * | 2018-05-10 | 2018-10-12 | 南京道熵信息技术有限公司 | Hard disk corruptions prediction technique and device based on SMART information and deep learning |
-
2018
- 2018-12-24 CN CN201811583676.2A patent/CN109857607A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964213A (en) * | 2009-07-24 | 2011-02-02 | 中芯国际集成电路制造(上海)有限公司 | Test method on failure analysis of storage cell |
| US20160104539A1 (en) * | 2014-10-08 | 2016-04-14 | Kyungryun Kim | Storage device and reliability verification method |
| CN107967928A (en) * | 2017-10-18 | 2018-04-27 | 武汉忆数存储技术有限公司 | A kind of flash chip life-span prediction method based on mathematical model |
| CN108647136A (en) * | 2018-05-10 | 2018-10-12 | 南京道熵信息技术有限公司 | Hard disk corruptions prediction technique and device based on SMART information and deep learning |
Non-Patent Citations (1)
| Title |
|---|
| R. KHUBCHANDANI: ""A Fast Test to Generate Flash Memory Threshold Voltage Distribution Map"", 《RECORDS OF THE 1999 IEEE INTERNATIONAL WORKSHOP ON MEMORY TECHNOLOGY, DESIGN AND TESTING》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110580932A (en) * | 2019-08-29 | 2019-12-17 | 华中科技大学 | Memory cell quality measurement method applied to wear leveling |
| CN110580932B (en) * | 2019-08-29 | 2021-04-06 | 华中科技大学 | Memory cell quality measurement method applied to wear leveling |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105808227B (en) | A kind of fine arts resource file detection method and device | |
| CN105629156B (en) | Analog circuit fault based on genetic planning tests optimal sequential search method | |
| US11036507B2 (en) | Processor testing using pairs of counter incrementing and branch instructions | |
| US10254336B2 (en) | Iterative N-detect based logic diagnostic technique | |
| CN104572321A (en) | Test selection method and device based on domination criterion | |
| CN109901996A (en) | Method, apparatus, electronic equipment and the readable storage medium storing program for executing of subtest | |
| CN104122497A (en) | Circuit and method for generating test vectors required by built-in self-test of integrated circuit | |
| JP2003007089A (en) | Method for restoring the state of the algorithm control when error occurs during testing device to be tested | |
| CN110082666A (en) | Chip testing analysis method, device, equipment and storage medium | |
| CN109857607A (en) | A kind of reliability checking method and device of NAND Flash solid state hard disk | |
| CN108122596A (en) | The test method and device of a kind of memory | |
| CN104063307A (en) | Software testing method and system | |
| CN113946983A (en) | Method and device for evaluating weak links of product reliability and computer equipment | |
| Pomeranz | A test selection procedure for improving the accuracy of defect diagnosis | |
| US9286181B2 (en) | Apparatus for capturing results of memory testing | |
| CN106503010A (en) | A kind of method and device of database change write subregion | |
| CN105487982B (en) | API method of testings and system | |
| US20140281719A1 (en) | Explaining excluding a test from a test suite | |
| CN105224649B (en) | A kind of data processing method and device | |
| CN108766527B (en) | Method and device for determining food calorie | |
| CN108028530B (en) | ESD detection device, integrated circuit and method applied to digital integrated electronic circuit | |
| US11138099B2 (en) | Method for testing software, and computing device and computer-readable storage medium thereof | |
| US20180238962A1 (en) | Structurally assisted functional test and diagnostics for integrated circuits | |
| CN110880182B (en) | Image segmentation model training method, image segmentation device and electronic equipment | |
| CN108763039B (en) | Service fault simulation method, device and equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |