CN107832012A - A kind of method that online mining flash memory system journey difference phenomenon optimization refreshes - Google Patents
A kind of method that online mining flash memory system journey difference phenomenon optimization refreshes Download PDFInfo
- Publication number
- CN107832012A CN107832012A CN201711068052.2A CN201711068052A CN107832012A CN 107832012 A CN107832012 A CN 107832012A CN 201711068052 A CN201711068052 A CN 201711068052A CN 107832012 A CN107832012 A CN 107832012A
- Authority
- CN
- China
- Prior art keywords
- holding time
- flash block
- flash
- grade
- data
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/064—Management of blocks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0679—Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Read Only Memory (AREA)
- Techniques For Improving Reliability Of Storages (AREA)
Abstract
The invention discloses the method that a kind of optimization of online mining flash memory system journey difference phenomenon refreshes.This method is intended to monitor the supported minimum refreshing frequency of each flash block in real time to reduce refreshing frequency.Traditional flushing policy is to select refreshing frequency according to erasing times for each flash block.And processing procedure difference shows that each flash block can support different erasing times, therefore traditional method for refreshing can cause many unnecessary refreshings.The present invention proposes the real-time monitoring real reliability of flash block to judge the flash block supported holding time, selects supported minimum refreshing frequency to reduce unnecessary refresh work with this.In addition, if the holding time of data is very short, refresh operation can be skipped.And then the present invention proposes to be distributed to the flash block that can provide the slightly larger holding time according to the holding time of data further to reduce refresh work.According to the above method, the refresh operation of redundancy will be reduced, so as to which the life-span of flash memory and performance will be significantly improved.
Description
Technical field
It is excellent more particularly to a kind of online mining flash memory system journey difference phenomenon the present invention relates to reliability of flash memory technical field
Change the method refreshed.
Background technology
The advantages that due to good random access performance, low-density, low-power consumption, the flash memory device based on flash memory is
Progressively substitute traditional magnetic disk, turn into important storage device.With the raising of the reduction density of flash memory size, flash memory is shown
New feature.First, the guarantee of reliability of flash memory becomes new challenge.Wherein, holding time mistake is cardinal error,
Be exactly for a long time preserve data read when it is possible that mistake.In order to ensure the integrality of data, refresh work is drawn
Enter.Secondly, processing procedure difference phenomenon becomes a kind of phenomenon of generally existing in flash memory.It shows that each flash block possesses in flash memory
Different restrains oneself ability.
Refreshing is to be re-writed new position when the data that long-time preserves will appear from mistake, therefore is one
The operation of kind redundancy.Traditional method for refreshing is to determine its refreshing frequency according to the erasing times of flash block.With flash memory
Abrasion, the erasing times of flash block are increasing, refreshing frequency just more and more higher.But processing procedure difference phenomenon shows that flash block can
Different erasing times are supported, therefore traditional method for refreshing can introduce many unnecessary refresh works.
In order to reduce the refresh operation of redundancy, it is contemplated that correspondingly being carried by excavating the true endurance of flash block
For refreshing frequency, to reduce unnecessary refreshing.In addition, when refresh operation occurs, valid data still be present in flash block,
Refresh operation can be just performed.Therefore, correspondingly being distributed according to the supported different refreshing frequencys of flash block needs different preservations
The data of time can further reduce refresh work.This invention can reduce the refresh operation of redundancy, so as to the life-span of flash memory
It can be got a promotion with performance.
The content of the invention
The invention provides the method that a kind of optimization of online mining flash memory system journey difference phenomenon refreshes.Based on flash memory system path difference
Different phenomenon, present invention optimizes in order to ensure the redundancy refresh operation of reliability of flash memory.
The technical problems to be solved by the invention realize that it comprises the following steps by such technical scheme：
The holding time { grade 1, grade 2 ..., grade N } that step 1, a series of different lifes of setting can use,
The holding time of its middle grade 1 is maximum, and grade N holding time is minimum.Different grades each safeguards a refreshing queue, team
The address of each flash block is recorded in row.By flash block by holding time grade classification into three regions, cold-zone domain, temperature area and
Thermal region.The flash block in cold-zone domain is the flash block that holding time grade is 1~grade of grade C, and the flash block of temperature area is
Holding time grade is grade C~grade H flash block, and the flash block of thermal region is that holding time grade is grade H~grade N
Flash block.During initialization, it is believed that all flash blocks can support the holding time of maximum, cold-zone domain is stored in, by its whole
Add the refreshing queue of grade 1.The data of arrival are all in the flash block of write-in grade 1.
Step 2, flash block all within grade i holding time must can with ensure flash memory by refresh all one time
By property.For the refresh operation triggered every time, read all Hash memory pages and calculate the most big page error bit number of flash block, and will have
Imitate data and write new Hash memory pages.When error bit number maximum between all Hash memory pages is more than a threshold value hr, grade i is selected
+ 1 as flash block current supported holding time；And when maximum error bit number is less than a threshold value hl, select grade
I-1 is as flash block current supported holding time；Otherwise, flash block holding time grade is constant.
Step 3, when the holding time grade of flash block changes, adjustment refresh queue, i.e., by the address of flash block
Move on in corresponding refreshing queue.
Step 4, when a write occurs the write, new data write-in cold-zone domain.But when renewal operation occurs for data, if data
Cold-zone domain is stored in before, then stores data into temperature area.If being stored in temperature area before data, store data into
Thermal region.Otherwise still thermal region is arrived in storage.When selecting the flash block in regional to be stored, each prioritizing selection can
Support the flash block of maximum holding time.
Step 5, when garbage collection operation is triggered, the holding time grade of flash block is reduced to next grade.And dodge
Valid data in counterfoil then need to move on to new position storage.If renewal operation occurs for more data on the contrary, being deposited before data
Store up in thermal region, be then stored to temperature area.If being stored in temperature area before data, cold-zone domain is stored to.It is no
Then still cold-zone domain is arrived in storage.When selecting the flash block in regional to be stored, each prioritizing selection can support maximum
The flash block of holding time.
The beneficial effects of the present invention are the flash block supported maximum holding time has been excavated, refresh frequency to reduce it
Rate.In addition, for refreshing principle, it would be desirable to further reduced in flash block corresponding to the data Cun Chudao of different holding times
Refresh operation.By this way, the refresh operation of flash memory internal redundancy is reduced, so as to improve its life-span and performance.
Brief description of the drawings
The brief description of the drawings of the present invention is as follows：
Fig. 1 is the flow chart of the present invention：
Fig. 2 is the holding time level diagram of the present invention：
Fig. 3 is the refreshing queue schematic diagram of the present invention：
Fig. 4 is the data and flash memory Block- matching assigning process schematic diagram of the present invention：
Embodiment
The invention will be further described with reference to the accompanying drawings and examples：
The present invention identifies refreshing frequency and distributes needs corresponding preservation by the true endurance of online mining flash block
The data of time reduce unnecessary refresh operation.In the process, its supported preservation is identified when flash block refreshes
Time.The holding time of insurance is selected for flash block and be mounted with effect data again in garbage reclamation.When data write, choosing
Suitable flash block is selected to be stored.
Fig. 1 is the flow chart of the present invention, since step 101, then：
In step 102, flash memory is initialized, sets a series of holding time grade { grade 1, grade 2 ..., grade N },
The holding time of its middle grade 1 is maximum, and grade N holding time is minimum.When initial identification flash block can support the preservation of grade 1
Between.Build the refreshing queue of grade 1.
In step 103, judge whether current each refreshing queue will perform refresh operation.If grade i needs to perform brush
New operation, then perform step 104.
All active pages in step 104, the flash block that will refresh re-write storage in new Hash memory pages and realize brush
New operation.
In step 105, read all pages in flash block and calculate the maximum error bit number in flash block.
In step 106, compare current flash block maximum error bit number and threshold value hr and hl size.If more than hr or
Step 107 is then performed less than hl, otherwise performs step 103.
It is the holding time grade that current flash block selects i+1 or i-1 in step 107.
In step 108, record current flash block holding time grade and update refreshing queue.
In step 109, the quantity of effective flash block in flash memory determines whether to trigger garbage reclamation mechanism, if
The quantity of effect flash block then performs step 110 less than a threshold value.
In step 110, valid data are judged whether, then need progress valid data to move if there is valid data
Move, then perform step 113.Otherwise step 111 is performed.
In step 111, judge whether current flash block has already been through the refresh work of a new round, if otherwise performing step
112。
In step 112, holding time grade of the i+1 grades for current flash block is selected, and continue executing with step 108.
In step 113, judge whether flash block belongs to cold-zone domain, if it is perform step 115, otherwise perform step
114。
In step 114, judge whether flash block belongs to temperature area, if it is perform step 116, otherwise perform step
115。
In step 115, active page is write into temperature area.
In step 116, active page is write into cold-zone domain.
In step 117, write operation occurs, if updating the data, then performs step 118, is otherwise then new data, and hold
Row step 120.
In step 118, judge whether legacy data is stored in thermal region, if it is perform step 121, otherwise perform step
119。
In step 119, judge whether legacy data is stored in temperature area, if it is perform step 121, otherwise perform step
122。
In step 120, the flash block in cold-zone domain is write data into.
In step 121, the flash block of temperature area is write data into.
In step 122, the flash block of thermal region is write data into.
When selecting the flash block of regional, prioritizing selection can support holding time most long flash block.
Fig. 2 is the holding time level diagram of the present invention：Select flash memory of a series of holding time for each period
Used in the block selection holding time, 1..N grade is shared, the holding time of its middle grade 1 is most long, and grade N holding time is most
It is short.The selection of this serial holding time can be selected by virtue of experience, according to demand all holding times can set, but one
Surely the holding time of grade 1 is met>The holding time of grade 2>…>Grade N holding time.In the holding time of flash block
On grade judges, this invention takes a series of effectively reliable criterions.When initial, the flash block supported holding time is most
It is long.When the maximum error bit number in flash block occurs more than hr or GC prior to refresh operation, next holding time etc. is selected
Holding time grade of the level as flash block.And when the maximum error bit number in flash block is less than hl, then select upper one to protect
Deposit the holding time grade that time grade is flash block.
Fig. 3 is the refreshing queue schematic diagram of the present invention.Each holding time grade safeguards a refreshing queue.According to each
The holding time of queue and the number setting time interval of flash block, refresh a flash block at regular intervals, refresh suitable
Sequence is since queue head.The flash block in all queues has been brushed when reaching the holding time.Need to safeguard in queue and dodge
The address information of counterfoil, to reach the purpose accurately refreshed.In addition, according to Fig. 2, the holding time grade of flash block may
Change.Need to update when the holding time grade of flash block changes to refresh queue, by the address information of flash block
Grade from before moves on to New r4 queue.Pay attention to, the flash memory block message insertion queue head of New r4 queue is added, so as to new
One wheel, which refreshes, preferentially refreshes such flash block.
Fig. 4 is the data and flash memory Block- matching assigning process schematic diagram of the present invention.The present invention is according to principle is refreshed, by data
The flash block of larger holding time can be provided by, which being stored in, can effectively reduce refreshing frequency.Shown according to this schematic diagram, flash memory quilt
It is divided into three parts, cold-zone domain, temperature area and thermal region.Wherein cold-zone domain is 1~grade of grade C flash block, so this kind of sudden strain of a muscle
The counterfoil holding time is most long.Temperature area is grade C~grade H flash block, and this kind of flash block holding time is longer.And thermal region
For grade H~grade N flash block, this kind of flash block can support the most short holding time.All write-in is cold for initial new data
Region.When data update, temperature area is arrived in renewal.And if the data of temperature area update again just writes thermal region.Work as heat
Thermal region is still write when the data in region update.And if when refresh operation or garbage collection operation occur still effectively
Data are present, then need to move to valid data into new Hash memory pages storage.In selection, if original data is present in hot-zone
Domain, then move to temperature area.And if data in temperature area, then move to cold-zone domain.If in cold-zone domain, continue to be stored in cold
Region.
Wherein, when selecting the flash block of regional, the flash block of maximum holding time can be supported in prioritizing selection region.
In addition, the division in these three regions is defaulted as 1/1/8, i.e. the flash block of cold-zone domain, temperature area and thermal region respectively accounts for
10%, 10% and the 80% of flash memory.Other rationally region divisions are also feasible.
Claims (10)
1. a kind of method that online mining flash memory system journey difference phenomenon optimization refreshes, it is characterized in that, comprise the following steps：
The each flash block of step 1, the ONLINE RECOGNITION supported holding time；
Step 2, the holding time according to each flash block, add and refresh queue and to set its refreshing frequency etc. to be refreshed；
Step 3, the renewal time according to data, store data into corresponding flash block.
2. the method that the online mining flash memory system journey difference phenomenon optimization according to the claims 1 refreshes, it is characterized in that：
In step 1, a series of holding time grade { grade 1 ..., grade N } is initially set for flash block in each life cycle
Middle selection, the holding time of grade 1 is most long, and grade N holding time is most short.
3. according to the method described in the claims 2, it is characterized in that：Traditional method for refreshing is according to the erasing times of flash block
The refreshing frequency of flash block is set, this frequency and holding time are inversely proportional.Under processing procedure difference phenomenon, this conventional method meeting
Produce many unnecessary refresh operations.But a series of this holding time can be set according to conventional method, can also be rule of thumb
Setting.
4. the method that the online mining flash memory system journey difference phenomenon optimization according to the claims 1 refreshes, it is characterized in that：
Flash block can support the selection of holding time to be judged according to most big page error bit number in flash block during refreshing.Work as maximum
When page error bit number is more than a threshold value hr, the holding time of lower level is selected as its supported holding time.Such as
When fruit most big page error bit number is less than a threshold value hl, when selecting the holding time of a upper grade as its supported preservation
Between.
5. according to the method described in the claims 4, it is characterized in that：If the current grade of flash block is 1, then is being selected
It is still grade 1 during a upper grade.And if the current grade of flash block is N, then think flash memory when selecting lower level
Block, which breaks down, to be continuing with.
6. the method that the online mining flash memory system journey difference phenomenon optimization according to the claims 1 refreshes, it is characterized in that：
In step 2, each holding time grade safeguards a refreshing queue, and queue needs to safeguard the address information of flash block.
7. according to the method described in the claims 4, it is characterized in that：Refresh operation can have a variety of by the way of.First
Kind is asynchronous refresh, and refresh operation performs since queue head, has refreshed all flash blocks when reaching the holding time.
In addition, the number of flash block is also required to safeguard, calculates refresh interval according to holding time and flash block number, and refresh frequency
Rate is inversely proportional.For second that demand refreshes, to the holding time after refresh whole flash blocks in queue.Other feasible refreshings
Operation is also admissible.
8. the method that the online mining flash memory system journey difference phenomenon optimization according to the claims 1 refreshes, it is characterized in that：
In step 3, the differentiation of holding time length, cold-zone domain, temperature area can be supported by flash memory is divided into three regions carrying out flash blocks
And thermal region.Wherein the flash block in cold-zone domain is 1~grade of grade C flash block, and the supported holding time is longer.Temperature area
Flash block be grade C~grade H flash block, the supported holding time is longer.And the flash block of thermal region be grade H~
Grade N flash block, supported holding time are shorter.New data is all initially write into cold-zone domain.When being preserved with data
Between judgement, progressively select suitable region to store.When selecting the flash block of regional, prioritizing selection can be supported most to grow guarantor
Deposit the free flash memory block of time.
9. according to the method described in the claims 8, it is characterized in that：Distribute what flash block used according to the holding time of data
Method is the holding time and flash block holding time of correction data.If data update before flash block refreshes, data
Holding time is less than the holding time of flash block.Otherwise the holding time of data is longer, it is necessary to which the sudden strain of a muscle of long holding time can be supported
Counterfoil stores.In addition, refresh for flash block needs the longer holding time with valid data during garbage reclamation, this kind of data
Flash block.
10. according to the method described in the claims 8, it is characterized in that：If the holding time that data need is longer, basis
The region stored before data reselects region.If thermal region, then temperature area is selected.If temperature area, then select
Thermal region.And if thermal region then continues to select thermal region.Vice versa for the data for possessing the shorter holding time, root during write-in
Region is also reselected according to the region stored before data.If cold-zone domain, then select temperature area.If temperature area, then
Select thermal region.And if thermal region, still select thermal region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711068052.2A CN107832012A (en) | 2017-11-03 | 2017-11-03 | A kind of method that online mining flash memory system journey difference phenomenon optimization refreshes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711068052.2A CN107832012A (en) | 2017-11-03 | 2017-11-03 | A kind of method that online mining flash memory system journey difference phenomenon optimization refreshes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107832012A true CN107832012A (en) | 2018-03-23 |
Family
ID=61651561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711068052.2A Pending CN107832012A (en) | 2017-11-03 | 2017-11-03 | A kind of method that online mining flash memory system journey difference phenomenon optimization refreshes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107832012A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109376094A (en) * | 2018-09-27 | 2019-02-22 | 郑州云海信息技术有限公司 | A kind of method of the cold and hot Data Migration of garbage reclamation in storage system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104575618A (en) * | 2015-01-13 | 2015-04-29 | 重庆大学 | Read-write modulation technique based on flash memory error checking |
| DE102012204256B4 (en) * | 2011-03-21 | 2015-08-20 | Denso Corporation | METHOD AND SYSTEM FOR INITIATING A RE-OPERATION PROCESS IN A NON-VOLATILE SOLID STORAGE DEVICE |
| CN105138472A (en) * | 2014-06-06 | 2015-12-09 | 华为技术有限公司 | Flash memory wear optimization method and device |
| CN105204958A (en) * | 2015-10-19 | 2015-12-30 | 哈尔滨工业大学 | Coding method for prolonging NAND Flash data reliable storage time |
| CN105242871A (en) * | 2014-06-06 | 2016-01-13 | 华为技术有限公司 | Data writing method and apparatus |
| CN105302474A (en) * | 2014-07-31 | 2016-02-03 | 华为技术有限公司 | Method for controlling write operation of flash memory and controller of flash memory |
| CN105632546A (en) * | 2015-07-21 | 2016-06-01 | 上海磁宇信息科技有限公司 | MRAM (Magnetic Random Access Memory) chip and self-refreshing operation method thereof |
| CN105679360A (en) * | 2016-01-05 | 2016-06-15 | 上海新储集成电路有限公司 | Refreshable nonvolatile memory and refreshing method thereof |
| CN106598484A (en) * | 2016-11-17 | 2017-04-26 | 华为技术有限公司 | Data storage method, flash memory chip and storage device |
| WO2017152392A1 (en) * | 2016-03-09 | 2017-09-14 | 华为技术有限公司 | Flash memory device refreshing method and apparatus |
-
2017
- 2017-11-03 CN CN201711068052.2A patent/CN107832012A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102012204256B4 (en) * | 2011-03-21 | 2015-08-20 | Denso Corporation | METHOD AND SYSTEM FOR INITIATING A RE-OPERATION PROCESS IN A NON-VOLATILE SOLID STORAGE DEVICE |
| CN105138472A (en) * | 2014-06-06 | 2015-12-09 | 华为技术有限公司 | Flash memory wear optimization method and device |
| CN105242871A (en) * | 2014-06-06 | 2016-01-13 | 华为技术有限公司 | Data writing method and apparatus |
| CN105302474A (en) * | 2014-07-31 | 2016-02-03 | 华为技术有限公司 | Method for controlling write operation of flash memory and controller of flash memory |
| CN104575618A (en) * | 2015-01-13 | 2015-04-29 | 重庆大学 | Read-write modulation technique based on flash memory error checking |
| CN105632546A (en) * | 2015-07-21 | 2016-06-01 | 上海磁宇信息科技有限公司 | MRAM (Magnetic Random Access Memory) chip and self-refreshing operation method thereof |
| CN105204958A (en) * | 2015-10-19 | 2015-12-30 | 哈尔滨工业大学 | Coding method for prolonging NAND Flash data reliable storage time |
| CN105679360A (en) * | 2016-01-05 | 2016-06-15 | 上海新储集成电路有限公司 | Refreshable nonvolatile memory and refreshing method thereof |
| WO2017152392A1 (en) * | 2016-03-09 | 2017-09-14 | 华为技术有限公司 | Flash memory device refreshing method and apparatus |
| CN106598484A (en) * | 2016-11-17 | 2017-04-26 | 华为技术有限公司 | Data storage method, flash memory chip and storage device |
Non-Patent Citations (1)
| Title |
|---|
| YEJIA DI: "Exploiting Process Variation for Retention Induced Refresh Minimization on Flash Memory", 《IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION SYSTEMS (TVLSI)》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109376094A (en) * | 2018-09-27 | 2019-02-22 | 郑州云海信息技术有限公司 | A kind of method of the cold and hot Data Migration of garbage reclamation in storage system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105242871B (en) | A kind of method for writing data and device | |
| CN104834608B (en) | A kind of buffer replacing method under isomery memory environment | |
| CN104834609B (en) | Multi-level buffer method based on history promotion and demotion frequency | |
| CN105094686B (en) | Data cache method, caching and computer system | |
| CN105005617B (en) | A kind of storage method and device of time series data | |
| CN102843396B (en) | Data write-in and read method and device in a kind of distributed cache system | |
| CN104516471B (en) | Method and device for managing power supply of storage system | |
| CN104699422B (en) | Data cached determination method and device | |
| CN106527988B (en) | A kind of method and device of solid state hard disk Data Migration | |
| CN104216665B (en) | The memory management method of multilevel-cell solid state hard disc | |
| CN104615594B (en) | A kind of data-updating method and device | |
| CN104798055B (en) | Memory Controller, data storage device and memory control methods | |
| US20140181432A1 (en) | Priority-based garbage collection for data storage systems | |
| CN103440207A (en) | Caching method and caching device | |
| CN103514110B (en) | The buffer memory management method of non-volatile memory device and device | |
| US20130339574A1 (en) | Variability aware wear leveling | |
| CN107391398B (en) | Management method and system for flash memory cache region | |
| CN104699424B (en) | A kind of isomery EMS memory management process based on page temperature | |
| ATE422688T1 (en) | STRATEGIES FOR READING INFORMATION FROM A MASS STORAGE MEDIUM USING A CACHE MEMORY | |
| CN103257935A (en) | Cache management method and application thereof | |
| CN108139872A (en) | A kind of buffer memory management method, cache controller and computer system | |
| CN105786410A (en) | Method for increasing processing speed of data storage system and data storage system | |
| CN111580754B (en) | Write-friendly flash memory solid-state disk cache management method | |
| CN106933516A (en) | A kind of storage controlling method and device of the storage system of blending agent | |
| CN102981977A (en) | Data storage method capable of prolonging service life of abode flash player (FLASH) |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180323 |