CN108108128A - A kind of abrasion equilibrium method and SSD for mixing SSD - Google Patents
A kind of abrasion equilibrium method and SSD for mixing SSD Download PDFInfo
- Publication number
- CN108108128A CN108108128A CN201711370571.4A CN201711370571A CN108108128A CN 108108128 A CN108108128 A CN 108108128A CN 201711370571 A CN201711370571 A CN 201711370571A CN 108108128 A CN108108128 A CN 108108128A
- Authority
- CN
- China
- Prior art keywords
- slc
- mlc
- data
- wear
- area
- 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
- G06F3/061—Improving I/O performance
-
- 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
-
- 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
- G06F3/0674—Disk device
- G06F3/0676—Magnetic disk device
Abstract
The invention discloses a kind of for mixing the abrasion equilibrium method and solid state disk of SSD, it is characterized in that including abrasion equilibrium in SLC area simultaneously, abrasion equilibrium and interregional abrasion equilibrium in MLC regions, abrasion equilibrium is by regarding SLC area as round-robin queue in SLC area, abrasion equilibrium is by counting the erasable number of each block in MLC regions, the few block of the erasable number of prioritizing selection is used to store data, the interregional abrasion equilibrium is to identify that the data in SLC are divided into cold data and dsc data by threshold values according to dsc data, cold data is moved into MLC regions from SLC area, the dsc data identification threshold values is dynamically determined according to rate of depreciation.Using SLC readwrite performances it is high and it is erasable often the advantages of store dsc data, cold data is stored using the MLC of large capacity, so as to reach the compromise of cost, capacity, service life, performance.The probability that cold data write-in SLC, dsc data are written into MLC is reduced by using dynamic adjustment dsc data threshold value, reduces the data volume that SLC is migrated to MLC.
Description
Technical Field
The invention relates to a solid state disk control technology, in particular to a wear leveling method of a hybrid SSD and the SSD.
Background
At present, as a Solid State Disk (SSD) shows a better characteristic than a conventional mechanical hard disk, the SSD has the advantages of high read/write performance, low power consumption, small size, and the like, and has gradually replaced the mechanical hard disk in a plurality of fields.
With the development of NANDflash technology, SSDs have shifted from using SLC (single level cell) to MLC (triple level cell). The disadvantage of relatively high SSD cost and short lifetime is caused by using all blocks in MLC grains as MLC, and it is a trend to open some blocks in MLC grains into SLC part into MLC. Because the erasing life of each Block of the MLC type is limited, in order to improve the whole service life of the MLC type, the MLC type must be subjected to wear leveling, the selection of a wear leveling strategy is very important, and if the wear leveling strategy is not good, the problems of write amplification and reduction of the service life of the MLC type can be caused.
Fig. 1 is a schematic diagram of a conventional MLC type SSD data equalization strategy, and an SSD using a single MLC storage medium has disadvantages of high cost and short lifetime. The method comprises the steps that part of an MLC is generally converted into an SLC type in a forced mode, and therefore an SSD has SLC regions and MLC regions; generally, a wear leveling strategy is adopted, a host informs an SSD that a new write-in command exists, write command data enters a host data filter, the data volume of the host write-in command is used as a threshold value to judge cold and Hot data, if the data volume is larger than the threshold value, the Hot data is judged to be written into an SLC, and if the data volume is smaller than the threshold value, the cold data is judged to be written into an MLC; performing address mapping after cold and hot data judgment, performing data mapping on hot data respectively according to data types to perform SLC mapping, and performing MLC mapping on cold data; after address mapping, data can be written into an SLC region or an MLC region, the data of which the filter judges hot data is written into an SLC region, cold data is written into an MLC region, and cold data in an SLC is transferred into an MLC when the SLC region reaches an expected set threshold or when a garbage collection operation GC is triggered, which causes a large amount of unnecessary write operation in an SLC and a large amount of data transfer from the SLC region to the MLC region, thereby inevitably bringing about a large amount of write amplification problems.
The conventional MLC garbage recovery process is as follows, SLC regions are divided into host data regions hot regions and cold region hot data regions, and data is preferentially written into the hot data regions; when the hot data area is started, copying the data to a rom data area; when the spam recovery is carried out on the rom data area, transferring effective data with the cycle period number reaching N times to the MLC data area according to the N chance principles, otherwise, remaining in the rom data area; the MLC data area is operated according to the normal garbage collection mechanism. On one hand, the existing MLC SSD can prolong the stay time of hot data in the SLC, and improve the performance of the SSD in the initial use stage; on the other hand, the mechanism increases the data migration times inside the SLC, increases the garbage collection expense and the abrasion degree of the SLC, and sometimes cannot be paid.
Disclosure of Invention
Aiming at the defects, the invention aims to reduce the probability of data reading and writing caused by balancing the SLC area data to the MLC area, thereby realizing the reduction of the problem of writing amplification.
The invention provides a wear leveling method of a hybrid SSD, which aims to solve the problems, the SSD converts part of MLC type storage areas into SLC type storage areas or separately sets an SLC FLASH as SLC areas, the SSD is divided into the MLC areas and the SLC areas, and the wear leveling method is characterized by simultaneously comprising SLC area internal wear leveling, MLC area internal wear leveling and inter-area wear leveling, the SLC area internal wear leveling realizes the leveling by using the SLC areas as a circular queue, the MLC area internal wear leveling counts the erasing times of each block, and preferentially selects the block with less erasing times for storing data, the inter-area wear leveling is to identify the data in the SLC threshold value into cold data and hot data according to the hot data, the cold data is transferred from the SLC areas to the MLC areas, and the hot data identification dynamically determines the threshold value according to the wear rate.
The wear leveling method of the hybrid SSD is characterized in that the wear rates of the SLC and the MLC are used for adjusting the hot data threshold value theta, N S Is the total effective block number, N, of the SLC region M is Total effective block number, L, of MLC region S For SLC erase cycle, L M For an erase period of MLC, E S i Is the number of times of erasing each block of SLC, E M j Is the erase times of each block of the MLC, then the relative wear rates of SLC and MLC RW S And RW M It is expressed as:
the equivalent total erase times are:when [ RW ] S ]＝[RW M ],[]For integer sign calculation, SLC and MLC wear rate are the same, and reach wear balance between areas when [ RW S ]>[RW M ]In the mean time, SLC wear is larger than MLC wear block by theta, and data is written into MLC as much as possible until [ RW ] S ]＝[RW M ](ii) a When [ RW) S ]<[RW M ]In this case, the MLC wear needs to be reduced by θ to write data into SLC until RW is reached S ]＝[RW M ]。
The wear leveling method of the hybrid SSD is characterized in that the SLC is used as a circular queue, newly written data is written from a block pointed to by the head of the queue, when the SLC free block is insufficient, garbage collection operation is started, and garbage collection starts from a block pointed to by the tail of the queue.
The wear leveling method of the hybrid SSD is characterized in that a cycle _ time field is added to each mapping item of an SLC address mapping table, the cycle period number of data staying in an SLC area is recorded, and when a page corresponding to a garbage collection target block is an effective page and circulates for one week, the cycle _ time is increased by one; finally, determining a data migration strategy according to the cycle _ time size and the SLC \ MLC wear rate, and migrating the data in the SLC to the MLC by using the N-time chance principle when the SLC wear rate is slower than the MLC; otherwise, the data in the SLC is migrated directly to the MLC.
A solid state disk is characterized in that an SSD converts part of an MLC type storage area into an SLC type storage area, the SSD is divided into an MLC area and an SLC area, wear leveling in the MLC area and wear leveling between the SLC area are included, wear leveling in the SLC area is achieved by using the SLC area as a circular queue, wear leveling in the MLC area is achieved by counting erasing times of each block, the blocks with low erasing times are selected preferentially to be used for storing data, wear leveling between the SLC area is achieved by dividing data in an SLC into cold data and hot data according to a hot data identification threshold value, the cold data are moved from the SLC area to the MLC area, and the hot data identification threshold value is dynamically determined according to a wear rate.
The solid state disk is characterized in that the wear rates of the SLC and the MLC are used for adjusting the thermal data threshold values theta, N S Is the total effective block number, N, of the SLC region M is Total effective block number, L, of MLC region S For SLC erase cycle, L M For an erase period of MLC, E S i Is the number of times of erasing each block of SLC, E M j Is the number of times each block of the MLC is erased, then the relative wear rates of SLC and MLC, RW S And RW M It is expressed as:
the equivalent total erase times are:when [ RW ] S ]＝[RW M ],[]For rounding symbol calculation, SLC and MLC wear rate are the same, so as to achieve wear balance between regions when [ RW S ]>[RW M ]In this case, SLC wear is larger than MLC wear, and θ is increased to write data into MLC as much as possible until [ RW ] S ]＝[RW M ](ii) a When [ RW) S ]<[RW M ]In time, the MLC wear needs to be reduced by theta compared with SLC wear blocks, so that data can be written into SLC to the extent of [ RW ] as much as possible S ]＝[RW M ]。
The solid state disk is characterized in that the SLC is used as a circular queue, newly written data are written from a block pointed to by the head of the queue, when the SLC free block is insufficient, garbage collection operation is started, and garbage collection starts from a block pointed to by the tail of the queue.
The solid state disk is characterized in that a cycle _ time field is added to each mapping item of an SLC address mapping table, the cycle period number of data staying in an SLC area is recorded, and when a page corresponding to a garbage collection target block is an effective page and circulates for one week, the cycle _ time is increased by one; finally, determining a data migration strategy according to the cycle _ time size and the SLC/MLC wear rate, and when the SLC wear rate is slower than the MLC, migrating the data in the SLC to the MLC by using the N-time opportunity principle; otherwise, the data in the SLC is migrated directly to the MLC.
The invention has the beneficial effects that: by introducing the scheme, the advantages of high SLC read-write performance and high erasable times are utilized to store hot data, and the large-capacity MLC is utilized to store cold data, so that the compromise of cost, capacity, service life and performance is achieved. The probability that cold data is written into SLC and hot data is written into MLC is reduced by dynamically adjusting a hot data threshold, and the data volume of SLC transferred to MLC is reduced by managing SLC through a FIFO scheduling mechanism.
Drawings
FIG. 1 is a schematic diagram of a prior art MLC type SSD data equalization strategy;
FIG. 2 is a schematic diagram of an improved MLC type SSD data equalization strategy;
FIG. 3 is a SLC garbage collection flow chart.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
The SSD converts part of MLC type storage areas into SLC type storage areas, the SSD is divided into MLC areas and SLC areas, data of the areas are balanced through a data distribution mechanism, the data distribution mechanism mainly comprises a wear leveling control module (ware leveling control) and a hot and cold data distinguishing module (hot data filter), the wear leveling control module simultaneously comprises SLC area wear leveling, MLC area wear leveling and inter-area wear leveling, SLC area wear leveling uses the SLC areas as a circular queue, and head and tail are respectively head pointers and tail pointers of the queue, and the MLC area wear leveling. The MLC data area is composed of user data and system data, the user data stores data written in by a host, the system data storage mapping table and the bad block table are used for storing the data by counting erasing times of each block, the blocks with the small erasing times are preferably selected for storing the data, the inter-area wear balance is that the data in the SLC are divided into cold data and hot data according to a hot data identification threshold value, the cold data are transferred to the MLC area from the SLC area, and the hot data identification threshold value is dynamically determined according to the wear rate.
Fig. 2 is a schematic diagram of an improved MLC type SSD data equalization strategy, and the write operation flow is as follows:
1) When the HOST has a write command, firstly calculating the wear rates of the SLC and the MLC through a ware level control module, and dynamically adjusting a thermal data threshold theta by using the wear rates;
2) Then the data enters a hot data filter module, data with the load larger than theta are classified into hot data, and data with the load smaller than theta are classified into cold data;
3) Hot data obtains an SLC physical address corresponding to the logical address through an SLC map, and cold data obtains an MLC physical address corresponding to the logical address through an MLC map;
4) And finally, writing the hot data into the SLC and the cold data into the MLC.
The GC process of the invention is as follows:
1) When the SLC free block is insufficient, triggering a GC mechanism, and taking the data written first as a target block of garbage recovery;
2) When the SLC wear rate is slower than the MLC, using the N-time-of-chance principle to transfer the data in the SLC to the MLC; otherwise, directly migrating the data in the SLC to the MLC; recovering the block of the SLC after the data migration is finished;
3) The MLC data area adopts a normal GC flow;
the wear leveling strategy of the technical scheme is implemented in the following way:
wear leveling includes SLC \ MLC inter-region wear leveling and intra-region wear leveling. The inter-area wear leveling refers to wear leveling between SLC NAND flash and MLC NAND flash, and the intra-area wear leveling refers to wear leveling inside SLC and MLC grains.
Wear leveling in the area: because the SLC is used as a circular queue, the wear balance of the SLC particles can be met; the MLC counts the erasing times of each block, preferentially selects the blocks with the small erasing times to store data, and the mode is used for ensuring the wear balance in the MLC.
Wear leveling between areas: threshold adjustments and migration of cold data from SLC to MLC are identified by hot data with wear leveling.
The thermal data identification threshold θ adjustment is specifically determined as follows:
the thermal data identification threshold θ is one of the keys of the present invention. If θ is too small, a large amount of data will be allocated to SLC, causing increased SLC wear rate and causing a large amount of data to migrate from SLC to MLC; if θ is too large, the data will be heavily distributed to MLCs, which will result in faster MLC wear rates and reduced SSD overall write performance.
Adjusting θ using wear rates of SLC and MLC, assuming N S And N M Total effective block number, L, for SLC and MLC respectively S And L M Erase periods of SLC and MLC, respectively, E S i And E M j The number of erase times of each block of SLC and MLC, respectively, then the relative wear rates of SLC and MLC RW S And RW M Is expressed as
The MLCSSD has an equivalent total erase count of
When [ RW ] S ]＝[RW M ]([·]Rounded symbols) can be considered as having the same SLC and MLC wear rates, achieving wear leveling between regions. When [ RW ] S ]>[RW M ]In the mean time, SLC wear is larger than MLC wear block by theta, and data is written into MLC as much as possible until [ RW ] S ]＝[RW M ]. When [ RW) S ]<[RW M ]In time, the MLC wear needs to be reduced by theta compared with SLC wear blocks, so that data can be written into SLC to the extent of [ RW ] as much as possible S ]＝[RW M ]。
Fig. 3 is a schematic diagram of SLC data management and data migration mechanism, specifically illustrating:
to achieve wear leveling in the SLC area, a FIFO mechanism is implemented in the SLC. And taking the SLC as a circular queue, storing newly written data into the block pointed by the head, starting garbage collection when the SLC free block is insufficient, collecting the block pointed by the tail, and moving the tail pointer forwards.
SLC to MLC data migration requires a combination of wear leveling and FIFO scheduling mechanisms. Adding a cycle _ time field to each mapping item of the SLC address mapping table, recording the cycle period number of data staying in the SLC area, and adding one to the cycle _ time when the corresponding page in the garbage collection target block is an effective page and circulates for one cycle; finally, determining a data migration strategy according to the cycle _ time size and the SLC \ MLC wear rate, and migrating the data in the SLC to the MLC by using the N-time chance principle when the SLC wear rate is slower than the MLC; otherwise, the data in the SLC is migrated directly to the MLC.
N chance principles are specified: when the SLC wear rate is slower than MLC, if cycle _ time = N (maximum number of cycles), migrating data from the block to which SLC tail points to MLC; otherwise, the tail block data is copied to the block pointed by the head, and the cycle _ time is increased by one, and a new round of circulation is carried out.
While the invention has been described with reference to a single embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A wear leveling method of a hybrid SSD is characterized in that an SLC FLASH is independently arranged on the SSD and is used as an SLC area, the SSD is divided into an MLC area and an SLC area, the wear leveling in the SLC area, the MLC area and the inter-area wear leveling are simultaneously included, the wear leveling in the SLC area realizes the leveling by using the SLC area as a circular queue, the wear leveling in the MLC area counts the erasing times of each b l ock, the b l ocks with less erasing times are preferentially selected for storing data, the inter-area wear leveling is to divide the data in the SLC area into cold data and hot data according to a hot data identification threshold value, the cold data is transferred to the MLC area from the SLC area, and the hot data identification threshold value is dynamically determined according to a wear rate.
2. The method of wear leveling in a hybrid SSD of claim 1, wherein the thermal data threshold θ, N is adjusted using the wear rate of the SLC and MLC S Is the total effective block number, N, of the SLC region M is Total effective block number, L, of MLC region S For SLC erase cycle, L M For an erase period of MLC, E S i Is the number of erasures of each bl ock of the SLC, E M j Is the number of erasures of each b l ock of the MLC, then the relative wear rates of SLC and MLC RW S And RW M It is expressed as:
the equivalent total erase times are:when [ RW) S ]＝[RW M ],[]For rounding symbol calculation, SLC and MLC wear rate are the same, so as to achieve wear balance between regions when [ RW S ]>[RW M ]In the mean time, SLC wear is larger than MLC wear block by theta, and data is written into MLC as much as possible until [ RW ] S ]＝[RW M ](ii) a When [ RW ] S ]<[RW M ]In time, the MLC wear needs to be reduced by theta compared with SLC wear blocks, so that data can be written into SLC to the extent of [ RW ] as much as possible S ]＝[RW M ]。
3. The wear leveling method for hybrid SSDs as claimed in claim 1, wherein SLC is used as a circular queue, the newly written data is written from the block pointed to by the head of the queue, when SLC free block is not enough, garbage collection is started, and garbage collection starts from the block pointed to by the tail of the queue.
4. The wear leveling method for hybrid SSD according to claim 3, wherein the SLC address mapping table adds cyc l e _ t ime field to each mapping entry, records the cycle period number of data staying in the SLC area, and adds one to cyc l e _ t ime when the corresponding page in the garbage collection target block is valid and cycles one week; finally, determining a data migration strategy according to the cyc l e _ t ime size and the SLC \ MLC wear rate, and migrating the data in the SLC to the MLC by using the N-time chance principle when the SLC wear rate is slower than the MLC; otherwise, the data in the SLC is migrated directly to the MLC.
5. A solid state disk is characterized in that an SSD converts a part of MLC type storage areas into SLC type storage areas, the SSD is divided into MLC areas and SLC areas and comprises SLC area internal wear balance, MLC area internal wear balance and inter-area wear balance, the SLC area internal wear balance achieves balance by using the SLC areas as a circular queue, the MLC area internal wear balance achieves balance by counting erasing times of each b l ock, the b l ocks with the small erasing times are preferentially selected for storing data, the inter-area wear balance achieves the effect that the data in the SLC areas are divided into cold data and hot data according to a hot data identification threshold, the cold data are transferred to the MLC areas from the SLC areas, and the hot data identification threshold is dynamically determined according to wear rates.
6. Solid state disk according to claim 5, characterized in that the thermal data thresholds θ, N are adjusted using the wear rate of SLC and MLC S Is the total effective block number, N, of the SLC region M is Total effective block number, L, of MLC region S For SLC erase cycle, L M For an erase period of MLC, E S i Is the number of erasures of each bl ock of the SLC, E M j Is the number of erasures of each b l ock of the MLC, then the relative wear rates of SLC and MLC RW S And RW M It is expressed as:
the equivalent total erase times are:when [ RW) S ]＝[RW M ],[]For rounding symbol calculation, SLC and MLC wear rate are the same, so as to achieve wear balance between regions when [ RW S ]>[RW M ]In this case, SLC wear is larger than MLC wear, and θ is increased to write data into MLC as much as possible until [ RW ] S ]＝[RW M ](ii) a When [ RW) S ]<[RW M ]In time, the MLC wear needs to be reduced by theta compared with SLC wear blocks, so that data can be written into SLC to the extent of [ RW ] as much as possible S ]＝[RW M ]。
7. The solid state disk of claim 5, wherein the SLC is configured as a circular queue, newly written data is written from a block pointed to by a head of the queue, and when the SLC has insufficient free blocks, a garbage collection operation is initiated, and garbage collection starts from a block pointed to by a tail of the queue.
8. The solid state disk of claim 7, wherein the SLC address mapping table adds cyc l e _ t ime field to each mapping entry, records the cycle period number of data staying in the SLC area, and adds one to cyc l e _ t ime when the corresponding page in the garbage collection target block is a valid page and cycles for one week; finally, determining a data migration strategy according to the cyc l e _ t ime size and the SLC \ MLC wear rate, and when the SLC wear rate is slower than the MLC, migrating the data in the SLC to the MLC by using the principle of N times of opportunity; otherwise, directly migrating the data in the SLC to the MLC.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711370571.4A CN108108128A (en) | 2017-12-19 | 2017-12-19 | A kind of abrasion equilibrium method and SSD for mixing SSD |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711370571.4A CN108108128A (en) | 2017-12-19 | 2017-12-19 | A kind of abrasion equilibrium method and SSD for mixing SSD |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108108128A true CN108108128A (en) | 2018-06-01 |
Family
ID=62210981
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711370571.4A Pending CN108108128A (en) | 2017-12-19 | 2017-12-19 | A kind of abrasion equilibrium method and SSD for mixing SSD |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108108128A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108920107A (en) * | 2018-07-13 | 2018-11-30 | 深圳忆联信息系统有限公司 | Screen method, apparatus, computer equipment and the storage medium of cold data |
| 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 |
| CN109684119A (en) * | 2018-12-06 | 2019-04-26 | 天津津航计算技术研究所 | Optimization method applied to NAND Flash wear leveling |
| CN109712662A (en) * | 2018-12-06 | 2019-05-03 | 天津津航计算技术研究所 | NAND Flash main control chip embedded with interim dynamic threshold wear leveling module |
| CN110688326A (en) * | 2019-09-23 | 2020-01-14 | 西安微电子技术研究所 | Flash memory wear leveling algorithm based on multiple thresholds |
| WO2020015126A1 (en) * | 2018-07-20 | 2020-01-23 | 江苏华存电子科技有限公司 | Portable multi-functional hand-operated cutter and usage method |
| CN111104045A (en) * | 2018-10-25 | 2020-05-05 | 深圳市中兴微电子技术有限公司 | Storage control method, device, equipment and computer storage medium |
| CN111241002A (en) * | 2018-11-29 | 2020-06-05 | 爱思开海力士有限公司 | Memory system and operating method thereof |
| CN111258924A (en) * | 2020-01-17 | 2020-06-09 | 中国科学院国家空间科学中心 | Mapping method based on satellite-borne solid-state storage system self-adaptive flash translation layer |
| CN111324303A (en) * | 2020-01-21 | 2020-06-23 | 深圳忆联信息系统有限公司 | SSD garbage recycling method and device, computer equipment and storage medium |
| CN112506445A (en) * | 2020-12-29 | 2021-03-16 | 杭州电子科技大学 | Partition proportion self-adaptive adjusting method of homogeneous hybrid solid state disk |
| CN112948398A (en) * | 2021-04-29 | 2021-06-11 | 电子科技大学 | Hierarchical storage system and method for cold and hot data |
| CN114327274A (en) * | 2021-12-29 | 2022-04-12 | 深圳忆联信息系统有限公司 | Mapping table loading checking method and device based on solid state disk and computer equipment |
| CN114442946A (en) * | 2022-01-06 | 2022-05-06 | 联芸科技（杭州）有限公司 | Physical block management method and solid state disk |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202472635U (en) * | 2012-03-23 | 2012-10-03 | 山东华芯半导体有限公司 | Flash memory wear leveling device |
| CN104572489A (en) * | 2013-10-23 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Wear leveling method and wear leveling device |
-
2017
- 2017-12-19 CN CN201711370571.4A patent/CN108108128A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202472635U (en) * | 2012-03-23 | 2012-10-03 | 山东华芯半导体有限公司 | Flash memory wear leveling device |
| CN104572489A (en) * | 2013-10-23 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Wear leveling method and wear leveling device |
Non-Patent Citations (1)
| Title |
|---|
| 姚英彪，王发宽: "具有磨损均衡意识的混合固态硬盘FTL算法", 《计算机学报》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108920107B (en) * | 2018-07-13 | 2022-02-01 | 深圳忆联信息系统有限公司 | Method and device for screening cold data, computer equipment and storage medium |
| CN108920107A (en) * | 2018-07-13 | 2018-11-30 | 深圳忆联信息系统有限公司 | Screen method, apparatus, computer equipment and the storage medium of cold data |
| WO2020015126A1 (en) * | 2018-07-20 | 2020-01-23 | 江苏华存电子科技有限公司 | Portable multi-functional hand-operated cutter and usage method |
| 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 |
| CN111104045A (en) * | 2018-10-25 | 2020-05-05 | 深圳市中兴微电子技术有限公司 | Storage control method, device, equipment and computer storage medium |
| CN111241002A (en) * | 2018-11-29 | 2020-06-05 | 爱思开海力士有限公司 | Memory system and operating method thereof |
| CN111241002B (en) * | 2018-11-29 | 2023-04-25 | 爱思开海力士有限公司 | Memory system and method of operating the same |
| CN109712662A (en) * | 2018-12-06 | 2019-05-03 | 天津津航计算技术研究所 | NAND Flash main control chip embedded with interim dynamic threshold wear leveling module |
| CN109684119A (en) * | 2018-12-06 | 2019-04-26 | 天津津航计算技术研究所 | Optimization method applied to NAND Flash wear leveling |
| CN110688326A (en) * | 2019-09-23 | 2020-01-14 | 西安微电子技术研究所 | Flash memory wear leveling algorithm based on multiple thresholds |
| CN111258924A (en) * | 2020-01-17 | 2020-06-09 | 中国科学院国家空间科学中心 | Mapping method based on satellite-borne solid-state storage system self-adaptive flash translation layer |
| CN111258924B (en) * | 2020-01-17 | 2021-06-08 | 中国科学院国家空间科学中心 | Mapping method based on satellite-borne solid-state storage system self-adaptive flash translation layer |
| CN111324303A (en) * | 2020-01-21 | 2020-06-23 | 深圳忆联信息系统有限公司 | SSD garbage recycling method and device, computer equipment and storage medium |
| CN112506445A (en) * | 2020-12-29 | 2021-03-16 | 杭州电子科技大学 | Partition proportion self-adaptive adjusting method of homogeneous hybrid solid state disk |
| CN112506445B (en) * | 2020-12-29 | 2022-05-20 | 杭州电子科技大学 | Partition proportion self-adaptive adjustment method for homogeneous hybrid solid state disk |
| CN112948398B (en) * | 2021-04-29 | 2023-02-24 | 电子科技大学 | Hierarchical storage system and method for cold and hot data |
| CN112948398A (en) * | 2021-04-29 | 2021-06-11 | 电子科技大学 | Hierarchical storage system and method for cold and hot data |
| CN114327274A (en) * | 2021-12-29 | 2022-04-12 | 深圳忆联信息系统有限公司 | Mapping table loading checking method and device based on solid state disk and computer equipment |
| CN114327274B (en) * | 2021-12-29 | 2023-07-04 | 深圳忆联信息系统有限公司 | Mapping table loading checking method and device based on solid state disk and computer equipment |
| CN114442946A (en) * | 2022-01-06 | 2022-05-06 | 联芸科技（杭州）有限公司 | Physical block management method and solid state disk |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108108128A (en) | A kind of abrasion equilibrium method and SSD for mixing SSD | |
| CN105487986B (en) | Sizing a cache memory to account for total byte write requirements | |
| US8103820B2 (en) | Wear leveling method and controller using the same | |
| US8171207B2 (en) | Adaptive hybrid density memory storage device and control method thereof | |
| KR101894625B1 (en) | Priority-based garbage collection for data storage systems | |
| CN102880556B (en) | Wear leveling method and system of Nand Flash | |
| US10503411B2 (en) | Data storage device and method for operating non-volatile memory | |
| US8015346B2 (en) | Memory system having hybrid density memory and methods for wear-leveling management and file distribution management thereof | |
| CN108255419A (en) | A kind of abrasion equilibrium method and SSD for TLC types SSD | |
| CN106293521B (en) | A kind of flash translation layer (FTL) management method that mapping granule is adaptive | |
| CN106874213B (en) | Solid state disk hot data identification method fusing multiple machine learning algorithms | |
| CN110413537B (en) | Flash translation layer facing hybrid solid state disk and conversion method | |
| US10346296B2 (en) | Data storage device | |
| TW201310454A (en) | Flash memory controller | |
| CN108829346B (en) | User writing request processing method of solid state disk adapting to flash memory page difference | |
| CN109471594B (en) | M L C flash memory read-write method | |
| CN105278875B (en) | A kind of mixing isomery NAND solid state hard disk | |
| US20120173796A1 (en) | Method for performing block management, and associated memory device and controller thereof | |
| CN106873912A (en) | The dynamic partition storage method and device, system of TLC chip solid state hard discs | |
| CN110688326A (en) | Flash memory wear leveling algorithm based on multiple thresholds | |
| CN107305478B (en) | Storage device, control unit thereof and data storage method for storage device | |
| CN112162695A (en) | Data caching method and device, electronic equipment and storage medium | |
| US20190012260A1 (en) | Flash memory package and storage system including flash memory package | |
| CN116457885A (en) | Data processing method, device and system | |
| CN111352593B (en) | Solid state disk data writing method for distinguishing fast writing from normal writing |
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: 20180601 |