CN109783024A - Data storage handling method and device - Google Patents
Data storage handling method and device Download PDFInfo
- Publication number
- CN109783024A CN109783024A CN201910019585.4A CN201910019585A CN109783024A CN 109783024 A CN109783024 A CN 109783024A CN 201910019585 A CN201910019585 A CN 201910019585A CN 109783024 A CN109783024 A CN 109783024A
- Authority
- CN
- China
- Prior art keywords
- characteristic
- data
- characteristic information
- information template
- feature
- 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
Abstract
The invention discloses a kind of data storage handling method and device, method is the following steps are included: set characteristic information template, and correspond to one unique ID number of association;Feature detection is carried out according to input data and write order, obtains the characteristic of input data;Characteristic and characteristic information template are compared one by one, and matched;If successful match, skips write operation and directly carry out return command completion information.It is detected by the feature to write-in data and reading data, simplify the read-write process with the data of specific rule, it is not write (Program) really to current data packet and reads (Read) operation, directly carry out data transmission, write operation and read time are saved, thus the write performance and reading performance of lifting system.
Description
Technical field
The present invention relates to field of data storage, a kind of data storage handling method and device are especially related to.
Background technique
Existing reading and writing data scheme is as follows: the user data for all preparations write-in memory bank that host sends over, main
Control chip is made no exception, and is not analyzed it, is only increased some check codes and additional control for error correction behind
Information, after then passing through logical address to the conversion of physical address, be written in memory bank (such as Nand Flash, Nor
Flash etc.).This processing method has problems in that: the memory space of needs is larger;Data write-in memory bank reads and deposits
Storage body requires to spend more time, influences system performance and efficiency.
All data are all uniformly processed in the prior art, and there is no carry out simplifying and speeding up place to those regular data
Reason, to influence whole efficiency.
Since in solid-state memory system, each write order operation and read command operation, most or most of times are all
It is to spend in the writing of memory bank itself (Program) or read in (Read) movement, data transmission only accounts for seldom one with data processing
Part-time.
Summary of the invention
In order to solve the defect of the above-mentioned prior art, the object of the present invention is to provide a kind of data storage handling method and dresses
It sets, for promoting read-write efficiency.
In order to achieve the above objectives, the technical scheme is that
A kind of data storage handling method, comprising the following steps:
Set characteristic information template, and one unique ID number of corresponding association;
Feature detection is carried out according to input data and write order, obtains the characteristic of input data;
Characteristic and characteristic information template are compared one by one, and matched;
If successful match, skips the direct return command of write operation and complete information.
Further, described to compare characteristic and characteristic information template one by one, and after carrying out matching step, packet
It includes,
If successful match, summary info is generated according to the corresponding ID number of characteristic information template and the logical address of write order;
Save summary info.
Further, described to skip after the direct return command of write operation completes information Step, including
Read command is obtained, corresponding summary info is searched according to the logical address in read command;
Judge that data to be read whether there is the characteristic information template to match according to the summary info;
If it exists, then corresponding characteristic is found according to characteristic information template;
Corresponding characteristic sequence, and backout feature data sequence are generated according to this feature data.
Further, the setting characteristic information template, and one unique ID number step of corresponding association, including,
It is 0 that the current ID number of all characteristic information templates, which is arranged,;
The content of initialization feature information model;
Unique corresponding ID number is associated with for the characteristic information template after initialization.
Further, the summary info includes for whether being matched with characteristic information template for representing characteristic
With mark.
Further, described that feature detection is carried out according to input data and write order, obtain the characteristic of input data
After step, including,
Obtain the frequency that different characteristic data occur;
Characteristic is arranged successively from high to low according to frequency of occurrence;
When the frequency of occurrence of characteristic is greater than preset value, newly-built ID number is associated with this feature data.
The present invention also proposes a kind of data storage processing device, comprising:
Feature templates setting unit, for setting characteristic information template, and one unique ID number of corresponding association;
Characteristic detection unit obtains input data for carrying out feature detection according to input data and write order
Characteristic;
Characteristic matching unit for comparing characteristic and characteristic information template one by one, and is matched;?
When with success, skips the direct return command of write operation and complete information.
Further, further includes:
Summary info generation unit is matched to the corresponding ID number of characteristic information template and write order logically for basis
Location generates summary info;
Summary info storage unit, for saving summary info.
It further, further include characteristic sequence generating unit, the characteristic sequence generating unit includes:
Searching module searches corresponding summary info according to the logical address in read command for obtaining read command;
Judgment module, for judging that data to be read whether there is the characteristic information mould to match according to the summary info
Plate;
Locating module, the characteristic for matching if it exists, then find corresponding feature according to characteristic information template
Data;
Read module, for generating corresponding characteristic sequence, and backout feature data sequence according to this feature data.
It further, further include AI processing unit, the AI processing unit includes:
The frequency obtains module, for obtaining the frequency of different characteristic data appearance;
Module is arranged, for being arranged successively characteristic from high to low according to frequency of occurrence;
Newly-built module, for when the frequency of occurrence of characteristic is greater than preset value, create ID number and this feature data into
Row association.
The beneficial effects of the present invention are: being detected by the feature to write-in data and reading data, simplifying has specific rule
The read-write process of the data of rule, current data packet is not write really (Program) and read (Read) operation, directly into
The transmission of row data, saves write operation and read time, thus the write performance and reading performance of lifting system.
Detailed description of the invention
Fig. 1 is a kind of method flow diagram of data storage handling method of one embodiment of the invention;
Fig. 2 is the method flow diagram that the present invention generates summary info;
Fig. 3 is the method flow diagram of initialization feature information model of the present invention;
Fig. 4 is the method flow diagram that the present invention creates that ID number is associated according to characteristic frequency of occurrence;
Fig. 5 present invention carries out the method flow diagram of reading data according to read command;
Fig. 6 is a kind of structural principle block diagram of data storage processing device of another embodiment of the present invention;
Fig. 7 is the structural block diagram of characteristic sequence generating unit of the invention;
Fig. 8 is the structural block diagram of AI processing unit of the invention.
Specific embodiment
To illustrate thought and purpose of the invention, the present invention is done further below in conjunction with the drawings and specific embodiments
Explanation.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiment is only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment, shall fall within the protection scope of the present invention.
It is to be appreciated that the directional instruction (up, down, left, right, before and after etc.) of institute is only used in the embodiment of the present invention
It explains in relative positional relationship, the motion conditions etc. under a certain particular pose (as shown in the picture) between each component, if the spy
When determining posture and changing, then directionality instruction also correspondingly changes correspondingly, and the connection, which can be, to be directly connected to, can also
To be to be indirectly connected with.
In addition, the description for being such as related to " first ", " second " in the present invention is used for description purposes only, and should not be understood as
Its relative importance of indication or suggestion or the quantity for implicitly indicating indicated technical characteristic.Define as a result, " first ",
The feature of " second " can explicitly or implicitly include at least one of the features.In addition, the technical side between each embodiment
Case can be combined with each other, but must be based on can be realized by those of ordinary skill in the art, when the combination of technical solution
Conflicting or cannot achieve when occur will be understood that the combination of this technical solution is not present, also not the present invention claims guarantor
Within the scope of shield.
Unless otherwise instructed, "/" herein represents meaning as "or".
Referring to Fig.1-5, one embodiment of the invention proposes a kind of data storage handling method, comprising the following steps:
S10, setting characteristic information template, and one unique ID number of corresponding association.
S20, feature detection is carried out according to input data and write order, obtains the characteristic of input data.
S30, characteristic and characteristic information template are compared one by one, and is matched.
If S40, successful match, skips the direct return command of write operation and complete information.
For step S10, characteristic information template is set and stored, the number of characteristic information template can be with one, can also be with
It is multiple.It when storing multiple template, is distinguished by the way that characteristic information template is associated with unique ID number one by one, need to only be searched
Rope ID number, can find corresponding characteristic information template can lead to after subsequent characteristics information model is associated with characteristic matching
ID number is crossed to position corresponding characteristic.
Wherein, the content of characteristic information template can be also possible in system operation dynamic with static initialization
Increase or update, once initialization is completed, each ID number forms association binding relationship with corresponding characteristic information template, passes through
ID number can index corresponding characteristic information template.
Specifically, step S10, including,
The current ID number of all characteristic information templates of S11, setting is 0.
The content of S12, initialization feature information model.
S13, unique corresponding ID number is associated with for the characteristic information template after initialization.
The purpose of initialization feature information model is, if having had been acknowledged one or more common feature sequences in advance
Column data, such as full 0 data, complete 1 data, then after initialization, it is established that feature letter corresponding with full 0 data or complete 1 data
Template is ceased, can be carried out simplified write-in and readout process since first group of data.This is to new solid state hard disk
Performance has an impact, and otherwise can just play promotion after subsequent continuous matching statistics generates corresponding characteristic information template
The effect of performance.
Either initial phase or subsequent dynamic update, characteristic information template for ID number, requiring is
It uniquely, between each other cannot be duplicate.That is: adjacent ID number, can be regular, such as can be continuous plus N number
Value, can also continuously subtract the numerical value of N, and wherein N is the natural number greater than 0, such as N is 1, then ID number can be 1,2,3,4 etc.;
It can also can follow without rule, will not only repeat, such as ID number is 2,31,55,6 etc..But regular ID number, is more advantageous to
The subsequent lookup of ID number and search, can be improved the search efficiency of summary info.
Initialization action can pass to solid state hard disk by special order by host, and solid state hard disk is receiving the life
When enabling, characteristic information template is initialized.Initialize how many a characteristic information templates and specific characteristic information template
Data are specified by this special order and are carried.The transmission channels of special command, can be PCIe/NVMe, can be string
Mouthful, it can be SPI Flash etc., here with no restrictions.The initial ID numbers of all characteristic information templates can be limited as 0,
It is exactly the state not initialized, is expressed as matching ID number in vain, ID number of the numerical value greater than zero is expressed as effective ID number.
For step S20, input data is obtained, and input data is analyzed, and then obtains the feature of input data
Data judge whether the writing process of input data can simplify by characteristic, by the characteristic for obtaining input data
According to come simplify input data write-in process, improve data write-in efficiency.
Under normal circumstances, input data is carried out in block transmission, such as solid hard disk as unit of certain length, number
According to usually being transmitted and handled as unit of 512 bytes, 4K byte or 16K byte.
Specifically, including: after step S20
S21, the frequency that different characteristic data occur is obtained.
S22, characteristic is arranged successively from high to low according to frequency of occurrence.
S23, when the frequency of occurrence of characteristic is greater than preset value, newly-built ID number is associated with this feature data.
The frequency that different characteristic data occur is obtained by being analyzed input data and being counted for step S21-S23
It is secondary, and characteristic is arranged successively record according to frequency of occurrence from high to low, when frequency of occurrence is higher than setting value, then represent
The input number of the input data of the type is relatively more, needs individually to establish a characteristic information template and is corresponding to it, to mention
The read-write efficiency of the input data of high the type, an and newly-built completely new ID number is corresponding to it association, while by new feature
Information model stores, in case subsequent calls.Certain ability of self-teaching is provided for system, in existing characteristic information mould
When plate is incomplete, play the role of augmented features information model.
Step S30 one by one compares the characteristic of input data with all existing characteristic information templates
It is right, such as comparison result and characteristic information template matching, then the corresponding ID number of this feature information model is recorded, for subsequent according to ID
Number production summary info.
Specifically, after step S30 further include:
If S31, successful match, abstract letter is generated according to the corresponding ID number of characteristic information template and the logical address of write order
Breath.
S32, summary info is saved.
For step S31 and S32, after according to the success of characteristic matching characteristic information model, ID number is carried out certain
The packing of rule handles to obtain summary info, and summary info generally comprises the logical address got from write order and (or simplifies
Logical address), match flag (1/0) and ID number.Specifically, match flag (1/0) represents whether matching succeeds using 1 and 0, such as
Fruit successful match match flag is 1, if matching is unsuccessful, match flag 0, therefore can be judged by match flag
Whether the data have in the presence of the characteristic template to match, and further call corresponding characteristic according to ID number.
Summary info is subjected to preservation and management, summary info can be stored with certain rule, with can be by patrolling
Volume address is found store its summary info for the purpose of.Such as: logical address is stored in into the mapping table of physical address.Specifically
, summary info can be deposited in the storage mediums such as DDR or Nand Flash.
For step S40, after successful match, flash memory storage body is no longer written in input data, but leaps to lower step,
To save the time of write operation;And other and all unmatched input data of all characteristic information templates, it is write by normal flow
Enter flash memory storage body, then carries out in next step.
After step S40, including data reading step:
S50, read command is obtained, corresponding summary info is searched according to the logical address in read command.
S60, judge that data to be read whether there is the characteristic information template to match according to the summary info.
S70, if it exists, then find corresponding characteristic according to characteristic information template.
S80, corresponding characteristic sequence, and backout feature data sequence are generated according to this feature data.
For step S50-S80, the read command sended over according to host obtains the logical address of data to be read, root
Summary info is found according to logical address, ID number corresponding with characteristic information template and match flag are contained in summary info, is led to
It crosses ID number and match flag obtains the data to be read and whether there is the characteristic information template and characteristic information template to match
Specifying information.
When determination exists and matches characteristic information template, specific characteristic is found for index with ID number, and with this
Characteristic is input, directly generates corresponding characteristic sequence, and backout feature data sequence that is to say data to be read,
Without reading data from NAND memory bank, the process of reading data is simplified, the efficiency of reading data is improved, works as generation
And after returning to the characteristic sequence of host sufficient length, returns to read command and complete information.
The data of characteristic sequence, that is, to be read, by the characteristic in characteristic information template, length is generated as needed.
The characteristic detected is the characteristic value detected from the burst of data of input, this characteristic value is stored in template;It reads
When out, whole string data is restored out using the characteristic value in this template.Such as: write operation is written a string to the address 0x1000
Number: 1234123412341234 ..., then, the characteristic value of this string number is exactly 1234, template is written in this characteristic value, and match
To a new ID number, such as 5.When host, which is issued, to be operated from 0x1000 reads, solid state hard disk is learnt by inquiry
There is characteristic matching in the address 0x1000, and corresponding ID number is 5, then generates feature sequence by the characteristic value 1234 represented by ID number is 5
Column 1234123412341234 ..., are directly returned to host.
This programme is detected by the feature to write-in data and reading data, simplifies the read-write with the data of specific rule
Process is not write (Program) really to current data packet and reads (Read) operation, directly carried out data transmission, save
Write operation and read time, thus the write performance and reading performance of lifting system.
With reference to Fig. 6-8, another embodiment of the present invention also proposes a kind of data storage processing device, comprising:
Feature templates setting unit 10, for setting characteristic information template, and one unique ID number of corresponding association.
Characteristic detection unit 20 obtains input data for carrying out feature detection according to input data and write order
Characteristic.
Characteristic matching unit 30 for comparing characteristic and characteristic information template one by one, and is matched, and
In successful match, skips the direct return command of write operation and complete information.
Summary info generation unit 40, for patrolling according to the corresponding ID number of characteristic information template and write order being matched to
It collects address and generates summary info.
Summary info storage unit 50, for saving summary info.
For feature templates setting unit 10, characteristic information template is set and stored, the number of characteristic information template can be with
One, it is also possible to multiple.When store it is multiple when template when, by by characteristic information template be associated with one by one unique ID number come into
Row is distinguished, and is only needed Searching I D, can be found corresponding characteristic information template, subsequent characteristics information model is matched with characteristic
After association, corresponding characteristic can be positioned by ID number.
Wherein, the content of characteristic information template can be also possible in system operation dynamic with static initialization
Increase or update, once initialization is completed, each ID number forms association binding relationship with corresponding characteristic information template, passes through
ID number can index corresponding characteristic information template.
Characteristic information template initialization procedure are as follows: it is 0 that the current ID number of all characteristic information templates, which is arranged,;Initialization
The content of ID number characteristic information template;Unique corresponding ID number is associated with for the characteristic information template after initialization.
The purpose of initialization feature information model is, if having had been acknowledged one or more common feature sequences in advance
Column data, such as full 0 data, complete 1 data, then after initialization, it is established that feature letter corresponding with full 0 data or complete 1 data
Template is ceased, can be carried out simplified write-in and readout process since first group of data.This is to new solid state hard disk
Performance has an impact, and otherwise can just play promotion after subsequent continuous matching statistics generates corresponding characteristic information template
The effect of performance.
Either initial phase or subsequent dynamic update, characteristic information template for ID number, requiring is
It uniquely, between each other cannot be duplicate.That is: adjacent ID number, can be regular, such as can be continuous plus N number
Value, can also continuously subtract the numerical value of N, and wherein N is the natural number greater than 0, such as N is 1, then ID number can be 1,2,3,4 etc.;
It can also can follow without rule, will not only repeat, such as ID number is 2,31,55,6 etc..But regular ID number, is more advantageous to
The subsequent lookup of ID number and search, can be improved the search efficiency of summary info.
Initialization action can pass to solid state hard disk by special order by host, and solid state hard disk is receiving the life
When enabling, characteristic information template is initialized.Initialize how many a characteristic information templates and specific characteristic information template
Data are specified by this special order and are carried.The transmission channels of special command, can be PCIe/NVMe, can be string
Mouthful, it can be SPI Flash etc., here with no restrictions.The initial ID numbers of all characteristic information templates can be limited as 0,
It is exactly the state not initialized, is expressed as matching ID number in vain, ID number of the numerical value greater than zero is expressed as effective ID number.
For characteristic detection unit 20, input data is obtained, and input data is analyzed, and then obtains input
The characteristic of data judges whether the writing process of input data can simplify by characteristic, inputs number by obtaining
According to characteristic come simplify input data write-in process, improve data write-in efficiency.
Under normal circumstances, input data is carried out in block transmission, such as solid hard disk as unit of certain length, number
According to usually being transmitted and handled as unit of 512 bytes, 4K byte or 16K byte.
For characteristic matching unit 30, the characteristic of input data, with all existing characteristic information templates
It is compared one by one, such as comparison result and characteristic information template matching, then records the corresponding ID number of this feature information model, be used for
It is subsequent that summary info is produced according to ID number.
After successful match, flash memory storage body is no longer written in input data, but leaps to lower step, writes behaviour to save
The time of work;And other and all unmatched input data of all characteristic information templates, flash memory storage body is written by normal flow,
It carries out again in next step.
For summary info generation unit 40, after according to the success of characteristic matching characteristic information model, by ID number into
The packing of the certain rule of row handles to obtain summary info, summary info generally comprise got from write order logical address (or
Simplified logical address), match flag (1/0) and ID number.Specifically, whether match flag (1/0) represents matching using 1 and 0
Success, if successful match match flag is 1, if matching is unsuccessful, match flag 0, therefore can be marked by matching
Will further calls corresponding characteristic according to ID number to judge whether the data have in the presence of the characteristic template to match
According to.
Summary info is carried out preservation and management by summary info storage unit 50, and summary info can be carried out with certain rule
Storage, for the purpose of it can be found by logical address and store its summary info.Such as: logical address is stored in physical address
Mapping table in.Specifically, summary info can be deposited in the storage mediums such as DDR or Nand Flash.
A kind of data storage processing device of the invention further includes characteristic sequence generation unit 60, characteristic sequence generation unit
60 include:
Searching module 61 searches corresponding summary info according to the logical address in read command for obtaining read command.
Judgment module 62, for judging that data to be read whether there is the characteristic information mould to match according to the summary info
Plate.
Locating module 63, the characteristic for matching if it exists, then find corresponding spy according to characteristic information template
Levy data.
Read module 64, for generating corresponding characteristic sequence, and backout feature data sequence according to this feature data
Column.
For characteristic sequence generation unit 60, the read command sended over according to host obtains the logic of data to be read
Summary info is found according to logical address in address, and ID number corresponding with characteristic information template, and matching are contained in summary info
Mark, obtaining the data to be read by ID number and match flag whether there is the characteristic information template and feature that match
The specifying information of information model.
When determination exists and matches characteristic information template, specific characteristic is found for index with ID number, and with this
Characteristic is input, directly generates corresponding characteristic sequence, and backout feature data sequence that is to say data to be read,
Without reading data from NAND memory bank, the process of reading data is simplified, the efficiency of reading data is improved, works as generation
And after returning to the characteristic sequence of host sufficient length, returns to read command and complete information.
The data of characteristic sequence, that is, to be read, by the characteristic in characteristic information template, length is generated as needed.
The characteristic detected is the characteristic value detected from the burst of data of input, this characteristic value is stored in template;It reads
When out, whole string data is restored out using the characteristic value in this template.Such as: write operation is written a string to the address 0x1000
Number: 1234123412341234 ..., then, the characteristic value of this string number is exactly 1234, template is written in this characteristic value, and match
To a new ID number, such as 5.When host, which is issued, to be operated from 0x1000 reads, solid state hard disk is learnt by inquiry
There is characteristic matching in the address 0x1000, and corresponding ID number is 5, then generates feature sequence by the characteristic value 1234 represented by ID number is 5
Column 1234123412341234 ..., are directly returned to host.
This programme is detected by the feature to write-in data and reading data, simplifies the read-write with the data of specific rule
Process is not write (Program) really to current data packet and reads (Read) operation, directly carried out data transmission, save
Write operation and read time, thus the write performance and reading performance of lifting system.
A kind of data storage processing device of the invention further includes AI processing unit 70, and the AI processing unit 70 includes:
The frequency obtains module 71, for obtaining the frequency of different characteristic data appearance.
Module 72 is arranged, for being arranged successively characteristic from high to low according to frequency of occurrence.
Newly-built module 73, for creating ID number and this feature data when the frequency of occurrence of characteristic is greater than preset value
It is associated.
AI processing unit 70 is by analyzing input data and being counted, the frequency that acquisition different characteristic data occur,
And characteristic is arranged successively record according to frequency of occurrence from high to low, when frequency of occurrence is higher than setting value, then representing should
The input number of the input data of type is relatively more, needs individually to establish a characteristic information template and is corresponding to it, to improve
The read-write efficiency of the input data of the type, an and newly-built completely new ID number is corresponding to it association, while by new feature letter
Breath template stores, in case subsequent calls.Certain ability of self-teaching is provided for system, in existing characteristic information template
When incomplete, play the role of augmented features information model.
This programme is detected by the feature to write-in data and reading data, simplifies the read-write with the data of specific rule
Process is not write (Program) really to current data packet and reads (Read) operation, directly carried out data transmission, save
Write operation and read time, thus the write performance and reading performance of lifting system.
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all utilizations
Equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content is applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of data storage handling method, which comprises the following steps:
Set characteristic information template, and one unique ID number of corresponding association;
Feature detection is carried out according to input data and write order, obtains the characteristic of input data;
Characteristic and characteristic information template are compared one by one, and matched;
If successful match, skips the direct return command of write operation and complete information.
2. data storage handling method as described in claim 1, which is characterized in that described by characteristic and characteristic information mould
Plate compares one by one, and after carrying out matching step, including,
If successful match, summary info is generated according to the corresponding ID number of characteristic information template and the logical address of write order;
Save summary info.
3. data storage handling method as claimed in claim 2, which is characterized in that described to skip the direct return command of write operation
After completing information Step, including
Read command is obtained, corresponding summary info is searched according to the logical address in read command;
Judge that data to be read whether there is the characteristic information template to match according to the summary info;
If it exists, then corresponding characteristic is found according to characteristic information template;
Corresponding characteristic sequence, and backout feature data sequence are generated according to this feature data.
4. data storage handling method as claimed in claim 2, which is characterized in that the setting characteristic information template, and it is right
It should be associated with a unique ID number step, including,
It is 0 that the current ID number of all characteristic information templates, which is arranged,;
The content of initialization feature information model;
Unique corresponding ID number is associated with for the characteristic information template after initialization.
5. data storage handling method as claimed in claim 2, which is characterized in that the summary info includes for representing
Whether characteristic is matched with the match flag of characteristic information template.
6. data storage handling method as described in claim 1, which is characterized in that it is described according to input data and write order into
Row feature detects, after the characteristic data step for obtaining input data, including,
Obtain the frequency that different characteristic data occur;
Characteristic is arranged successively from high to low according to frequency of occurrence;
When the frequency of occurrence of characteristic is greater than preset value, newly-built ID number is associated with this feature data.
7. a kind of data storage processing device characterized by comprising
Feature templates setting unit, for setting characteristic information template, and one unique ID number of corresponding association;
Characteristic detection unit obtains the feature of input data for carrying out feature detection according to input data and write order
Data;
Characteristic matching unit for comparing characteristic and characteristic information template one by one, and is matched, matching at
When function, skips the direct return command of write operation and complete information.
8. data storage processing device as claimed in claim 7, which is characterized in that further include:
Summary info generation unit, for raw according to the logical address for being matched to the corresponding ID number of characteristic information template and write order
At summary info;
Summary info storage unit, for saving summary info.
9. data storage processing device as claimed in claim 8, which is characterized in that further include that characteristic sequence generates list
Member, the characteristic sequence generating unit include:
Searching module searches corresponding summary info according to the logical address in read command for obtaining read command;
Judgment module, for judging that data to be read whether there is the characteristic information template to match according to the summary info;
Locating module, the characteristic for matching if it exists, then find corresponding characteristic according to characteristic information template;
Read module, for generating corresponding characteristic sequence, and backout feature data sequence according to this feature data.
10. data storage processing device as claimed in claim 7, which is characterized in that further include AI processing unit, at the AI
Managing unit includes:
The frequency obtains module, for obtaining the frequency of different characteristic data appearance;
Module is arranged, for being arranged successively characteristic from high to low according to frequency of occurrence;
Newly-built module, for when the frequency of occurrence of characteristic is greater than preset value, newly-built ID number to be closed with this feature data
Connection.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910019585.4A CN109783024B (en) | 2019-01-09 | 2019-01-09 | Data storage processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910019585.4A CN109783024B (en) | 2019-01-09 | 2019-01-09 | Data storage processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109783024A true CN109783024A (en) | 2019-05-21 |
CN109783024B CN109783024B (en) | 2022-02-01 |
Family
ID=66500275
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910019585.4A Active CN109783024B (en) | 2019-01-09 | 2019-01-09 | Data storage processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109783024B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110175000A (en) * | 2019-05-24 | 2019-08-27 | 深圳忆联信息系统有限公司 | Readwrite performance method for improving, device and computer equipment based on solid state hard disk |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000339104A (en) * | 1999-03-19 | 2000-12-08 | Hitachi Ltd | Disk sub-system and data copying method |
CN103544116A (en) * | 2012-07-09 | 2014-01-29 | 安凯(广州)微电子技术有限公司 | Data processing method and device |
CN106250785A (en) * | 2016-08-17 | 2016-12-21 | 成都市和平科技有限责任公司 | The guard method of a kind of data interruption transmission and data storage based on this guard method |
CN106294215A (en) * | 2015-06-24 | 2017-01-04 | 三星电子株式会社 | Storage device and the method for operating memory device |
CN106776190A (en) * | 2016-11-11 | 2017-05-31 | 北京京存技术有限公司 | A kind of eMMC writes method of testing and device |
-
2019
- 2019-01-09 CN CN201910019585.4A patent/CN109783024B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000339104A (en) * | 1999-03-19 | 2000-12-08 | Hitachi Ltd | Disk sub-system and data copying method |
CN103544116A (en) * | 2012-07-09 | 2014-01-29 | 安凯(广州)微电子技术有限公司 | Data processing method and device |
CN106294215A (en) * | 2015-06-24 | 2017-01-04 | 三星电子株式会社 | Storage device and the method for operating memory device |
CN106250785A (en) * | 2016-08-17 | 2016-12-21 | 成都市和平科技有限责任公司 | The guard method of a kind of data interruption transmission and data storage based on this guard method |
CN106776190A (en) * | 2016-11-11 | 2017-05-31 | 北京京存技术有限公司 | A kind of eMMC writes method of testing and device |
Non-Patent Citations (2)
Title |
---|
BIN DONG: "Data Elevator: Low-Contention Data Movement in Hierarchical Storage System", 《2016 IEEE 23RD INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING (HIPC)》 * |
董其文: "基于HDFS的小文件存储方法的研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110175000A (en) * | 2019-05-24 | 2019-08-27 | 深圳忆联信息系统有限公司 | Readwrite performance method for improving, device and computer equipment based on solid state hard disk |
Also Published As
Publication number | Publication date |
---|---|
CN109783024B (en) | 2022-02-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103455432B (en) | It is used for managing the method for a memory and its relevant memory | |
CN103345469B (en) | The storage of set of numbers, querying method and device thereof | |
US11269902B2 (en) | Time series data management method, device, and apparatus | |
CN102024046B (en) | Data repeatability checking method and device as well as system | |
US8176100B2 (en) | System for storing and managing objects | |
CN103049390A (en) | Applied metadata processing method and storing system | |
CN109669635B (en) | Command LBA (logical block addressing) coincidence detection method and device for solid-state storage master control chip | |
CN110168532A (en) | Data-updating method and storage device | |
CN107273046A (en) | A kind of data processing method and system based on Solid-state disc array | |
US20080162591A1 (en) | Method of Logging Transactions and a Method of Reversing a Transaction | |
CN104317850A (en) | Data processing method and device | |
US20150378609A1 (en) | Method for initializing nand flash | |
CN109783024A (en) | Data storage handling method and device | |
CN106528441B (en) | Data processing method and device for simulating EEPROM (electrically erasable programmable read-Only memory) and electronic equipment | |
CN106484719A (en) | A kind of method of extended mobile phone storage and terminal | |
CN110209349B (en) | Data processing method and terminal equipment | |
CN110288340B (en) | Data set storage method based on Java intelligent contract | |
CN103970683A (en) | Method for accelerating starting of USB flash disk | |
CN103605479B (en) | Data file wiring method and system, data file read method and system | |
CN110308941A (en) | BIOS compatibility method, device, terminal and computer storage medium without EDID screen | |
CN102117293B (en) | Dynamic file positioning and query method | |
CN106547472A (en) | Storage array management method and device | |
CN107908500A (en) | A kind of bad block identifying method and device | |
CN109634534B (en) | Method for quickly judging capacity of memory chip | |
CN104239236A (en) | Translation lookaside buffer and method for handling translation lookaside buffer deficiency |
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 |