CN105808156A - Method for writing data into solid state drive and solid state drive - Google Patents

Method for writing data into solid state drive and solid state drive Download PDF

Info

Publication number
CN105808156A
CN105808156A CN201410852946.0A CN201410852946A CN105808156A CN 105808156 A CN105808156 A CN 105808156A CN 201410852946 A CN201410852946 A CN 201410852946A CN 105808156 A CN105808156 A CN 105808156A
Authority
CN
China
Prior art keywords
write data
write
page
nandflash
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.)
Granted
Application number
CN201410852946.0A
Other languages
Chinese (zh)
Other versions
CN105808156B (en
Inventor
吴黎明
黄斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201410852946.0A priority Critical patent/CN105808156B/en
Priority to CN202010311530.3A priority patent/CN111638852A/en
Priority to PCT/CN2015/098109 priority patent/WO2016107442A1/en
Publication of CN105808156A publication Critical patent/CN105808156A/en
Application granted granted Critical
Publication of CN105808156B publication Critical patent/CN105808156B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input 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
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers

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)
  • Techniques For Improving Reliability Of Storages (AREA)

Abstract

Embodiments of the invention provide a method for writing data into a solid state drive (SSD) and the SSD. The SSD comprises a controller and a memory, wherein a storage medium of the memory comprises an NAND Flash and a non-volatile memory (NVM), and the NVM comprises a write region. The method comprises the steps of receiving a write operation request, wherein the write operation request comprises pre-written data; determining that the pre-written data is not stored in the write region; and writing the pre-written data into the write region. According to the embodiments of the invention, the pre-written data is written in the write region of the NVM, so that the accesses to the NAND Flash can be reduced, namely, the frequency of accessing the NAND Flash in the SSD is reduced, and the abrasion of the NAND Flash is reduced; moreover, due to the non-volatile characteristic of the NVM, data stored in the NVM is not lost under power failure, so that during power interruption, the SSD does not need to be powered by capacitors; and based on the reasons, the service life of the SSD is prolonged.

Description

Write data into method and the solid state hard disc of solid state hard disc
Technical field
The present embodiments relate to memory technology, particularly relate to a kind of method writing data into solid state hard disc and solid state hard disc.
Background technology
(SolidStateDrives, is called for short: SSD) solid state hard disc, is called for short solid dish, is widely used in server, desktop computer, notebook, mobile equipment, game machine etc..
As it is shown in figure 1, SSD10 many employings semiconductor N ANDFlash14 is as the storage medium of memorizer;And provide standby electricity by electric capacity 15, to avoid the data in internal memory 13 to lose when power down;By the read-write operation request of controller 12 Receiving Host 11, and internal memory 13 or NANDFlash14 are written and read.
In prior art, when read-write operation request is for NANDFlash14, by internal memory 13 as intermediary, data corresponding to read operation request need to be read to internal memory 13 from NANDFlash14, then return again to main frame 11;Or, data corresponding for write operation requests are first write internal memory 13, are then transferred to NANDFlash14 again.So, if data corresponding to the write operation requests being buffered in internal memory 13 are not also transferred to NANDFlash14, power failure, it is necessary to electric capacity 15 provides standby electricity, and the data in internal memory 13 are write NANDFlash14.
But, to add up based on historical data, NANDFlash and electric capacity are the major failure sources of SSD.Therefore, in SSD, the existence of electric capacity and that the read-write that NANDFlash is too much may result in service life of SSD is shorter.
Summary of the invention
The embodiment of the present invention provides a kind of method writing data into solid state hard disc and solid state hard disc, to increase the service life of SSD, promotes the reliability of SSD.
First aspect, the embodiment of the present invention provides a kind of method writing data into solid-state hard disk SSD, and described SSD includes controller and memorizer, and the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM includes writing region, and described method includes:
Receiving write operation requests, described write operation requests includes pre-write data;
Determine that described pre-write data write region described in being not stored in;
Region is write described in described pre-write data being write.
According to first aspect, in the first possible implementation of first aspect, described method also includes:
Judge whether described pre-write data have been stored in described NANDFlash;
If described pre-write data have been stored in described NANDFlash, it is invalid to be set to by the page storing described pre-write data in NANDFlash.
The first possible implementation according to first aspect or first aspect, in the implementation that the second of first aspect is possible, described by described pre-write data write described in write region after, described method also includes:
Preserving mapping relations, described mapping relations include the corresponding relation writing between the position in region described in the logical address of described pre-write data and the write of described pre-write data.
According to first aspect, first aspect the first in the possible implementation of the second any one, in the third possible implementation of first aspect, described method also includes:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
According to first aspect, first aspect the first in the third possible implementation any one, in the 4th kind of possible implementation of first aspect, described method also includes:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
Second aspect, the embodiment of the present invention provides a kind of method writing data into solid-state hard disk SSD, and described SSD includes controller and memorizer, and the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM includes writing region and reading field, and described method includes:
Receiving write operation requests, described write operation requests includes pre-write data;
Determine that described pre-write data have been stored in described reading field;
By the described pre-write Data Migration in described reading field to described in write region.
According to second aspect, in the first possible implementation of second aspect, described by the described pre-write Data Migration in described reading field to described in write region after, described method also includes:
Updating the first mapping relations is the second mapping relations, described first mapping relations include logical address and the described pre-write data corresponding relation between the position of described reading field of described pre-write data, and described second mapping relations include logical address and the described pre-write data corresponding relation between the described position writing region of described pre-write data.
The first possible implementation according to second aspect or second aspect, in the implementation that the second of second aspect is possible, described method also includes:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
According to second aspect, second aspect the first in the possible implementation of the second any one, in the third possible implementation of second aspect, described method also includes:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
According to second aspect, second aspect the first in the third possible implementation any one, in the 4th kind of possible implementation of second aspect, described NVM also includes clear area, described by the described pre-write Data Migration in described reading field to described in write region after, described method also includes:
In described clear area, adjusting N page to described reading field from the page writing region described in keeping for, described N is the number of pages that described pre-write data are corresponding.
The third aspect, the embodiment of the present invention provides a kind of solid-state hard disk SSD, and described SSD includes controller and memorizer, and the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, and described NVM includes writing region, wherein,
Described controller, is used for receiving write operation requests, and described write operation requests includes pre-write data;Determine that described pre-write data write region described in being not stored in;And, write region described in described pre-write data being write;
Described region of writing is for storing described pre-write data.
According to the third aspect, in the first possible implementation of the third aspect, described controller is additionally operable to:
Judge whether described pre-write data have been stored in described NANDFlash;
If described pre-write data have been stored in described NANDFlash, it is invalid to be set to by the page storing described pre-write data in NANDFlash.
The first possible implementation according to the third aspect or the third aspect, in the implementation that the second of the third aspect is possible, described controller is additionally operable to:
Preserving mapping relations, described mapping relations include the corresponding relation writing between the position in region described in the logical address of pre-write data and the write of described pre-write data.
According to the third aspect, the third aspect the first in the possible implementation of the second any one, in the third possible implementation of the third aspect, described controller is additionally operable to:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
According to the third aspect, the third aspect the first in the third possible implementation any one, in the 4th kind of possible implementation of the third aspect, described controller is additionally operable to:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
Fourth aspect, the embodiment of the present invention provides a kind of solid-state hard disk SSD, and described SSD includes controller and memorizer, and the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM and includes writing region and reading field, wherein,
Described controller, is used for receiving write operation requests, and described write operation requests includes pre-write data;Determine that described pre-write data have been stored in described reading field;And, by the described pre-write Data Migration in described reading field to described in write region.
According to fourth aspect, in the first possible implementation of fourth aspect, described controller is additionally operable to:
Updating the first mapping relations is the second mapping relations, described first mapping relations include logical address and the described pre-write data corresponding relation between the position of described reading field of described pre-write data, and described second mapping relations include logical address and the described pre-write data corresponding relation between the described position writing region of described pre-write data.
The first possible implementation according to fourth aspect or fourth aspect, in the implementation that the second of fourth aspect is possible, described controller is additionally operable to:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
According to fourth aspect, fourth aspect the first in the possible implementation of the second any one, in the third possible implementation of fourth aspect, described controller is additionally operable to:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
According to fourth aspect, fourth aspect the first in the third possible implementation any one, in the 4th kind of possible implementation of fourth aspect, described NVM also includes clear area, and described controller is additionally operable to:
In described clear area, adjusting N page to described reading field from the page writing region described in keeping for, described N is the number of pages that described pre-write data are corresponding.
The embodiment of the present invention provides a kind of method writing data into solid state hard disc and solid state hard disc, what pre-write data write NVM writes region, it is thus possible to reduce, NANDFlash is accessed, namely reduce the accessed frequency of NANDFlash in SSD, reduce the abrasion to NANDFlash;And, due to the non-volatile characteristic of NVM, the data power down being stored in NVM is not lost, and therefore, when power failure, SSD provides standby electricity without electric capacity.For these reasons, increase the service life of embodiment of the present invention SSD.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, introduce the accompanying drawing used required in embodiment or description of the prior art is done one simply below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the application scenarios schematic diagram of prior art SSD;
Fig. 2 by the memory element of SLC the schematic diagram of storage data;
Fig. 3 by the memory element of MLC the schematic diagram of storage data;
Fig. 4 is the structural representation of SSD of the present invention;
Fig. 5 is the topology example figure of NVM in SSD of the present invention;
Fig. 6 is the flow chart that the present invention writes data into the embodiment of the method one of SSD;
Fig. 7 is the flow chart that the present invention writes data into the embodiment of the method two of SSD;
Fig. 8 is mapping relations exemplary plot in the embodiment of the present invention.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.
In prior art, NANDFlash is the physical medium of SSD actual storage data.Wherein, NANDFlash includes single-layer type unit and stores (English: SingleLevelCell, be called for short: SLC), multiple field unit store (English: MultiLevelCell, it is called for short: MLC), (English: TripleLevelCell is called for short: TLC) several the storage of three-layer type unit.In enterprise-level field of storage, mainly use SLC and MLC.The erasable number of times of this medium of TLC is relatively low, is currently mainly used on the electronic product of consumer level.
The oxide film that SLC technical characterstic is among floating grid with source electrode is thinner, when write data by the electric charge making alive to floating grid, then pass through source electrode, stored electric charge can be eliminated, by such mode, just can storing 1 information unit, this technology is provided that quick programming and reading.But this technology is limited to silicon efficiency, and (English: Siliconefficiency), palpus, with more advanced flow process reinforcement technique, just can be lifted up SLC treatment technology.
It is (English: data bit), the data of storage are " 0 " or " 1 " is judged by based on voltage threshold that SLC stores 1 bit in each memory element.For the write of NANDFlash, namely program, it is simply that to its charging so that the voltage of NANDFlash is more than the voltage decision-point A shown in Fig. 2, and memory element means that " 0 ", namely written into;Without charging or voltage threshold lower than voltage decision-point A, mean that " 1 ", namely wipe, transverse axis representative voltage in Fig. 2.
Intel company succeeds in developing MLC at first in JIUYUE, 1997, its effect is that the information of two units is stored in a floating grid (English: FloatingGate), then utilize the electric charge of different potentials, precisely read and write by the Control of Voltage of memory storage.MLC is by using substantial amounts of electric pressure, and each unit stores two bits, and packing density is relatively larger.SLC stores 0 and 1 two value, and MLC can once store the value of more than 4.Therefore, MLC has the storage density that comparison is high.
As it is shown on figure 3, MLC stores the data of 2bit in each memory element, the data of storage are " 00 ", or " 01 ", or " 10 ", or " 11 " are also based on the judgement of voltage threshold.When voltage is less than voltage decision-point B, represent " 11 ";When voltage is between voltage decision-point B and voltage decision-point C, then represent " 10 ";Voltage is between voltage decision-point C and voltage decision-point D, then it represents that " and 01 ";And voltage reaches voltage more than decision-point D, then it represents that " 00 ".
One memory element of TLC can store 3bit, it is therefore desirable to the current potential of 8 grades is encoded decoding and could realize.Substantially, TLC belongs to MLC.
Current industry is in the ascendant about the research of new storage medium, and various new technology or novel storage medium emerge in an endless stream.New storage medium such as has phase transition internal memory (English: PhaseChangeRandomAccessMemory, be called for short: PCRAM/PCM), magnetic-resistance random access memory (English: MagnetoresistiveRandomAccessMemory, be called for short: MRAM), resistive formula random access memory (English: ResistiveRandomAccessMemory, be called for short: RRAM/ReRAM) and spin transfer torque random access memory (English: spintransfertorquerandomaccessmemory, being called for short: STT-RAM) etc., the present invention is restriction not.
These several novel storage mediums above-mentioned, (English: non-volatilememory is called for short: NVM) to broadly fall into non-volatile memory medium.In foreseeable future, based on the consideration on protection existing investment and cost, NANDFlash as primary storage medium by long-term existence.Meanwhile, novel non-volatile storage medium will progressively take on key player, until progressively replacing NANDFlash.In this evolution process, it is understood that there may be same SSD has the situation that medium coexists.
For these reasons, the embodiment of the present invention provides a kind of SSD.
As shown in Figure 4, SSD40 includes controller 42 and memorizer, and the storage medium of memorizer includes NANDFlash44 and NVM43.
Wherein, NVM and common memory the difference is that, NVM medium itself possesses power down and does not lose the characteristic of data.Based on this characteristic, NVM possesses following functional characteristic: buffer memory is write, and gathers into complete page in NVM, judges whether that transcription is to NANDFlash further according to the accessed frequency of this page and the amount of capacity of NVM;Add fast reading, by Data Migration high for accessed frequency to NVM.
Fig. 5 illustrates that the region of NVM divides.As it is shown in figure 5, NVM50 can include 4 logic regions: metadata area 51, reading field 52, write region 53 and clear area 54, it is achieved the partition management to NVM.The effect in each region is as follows:
Metadata area 51, for depositing the metadata of SSD.Such as, (FirstInputFirstOutput is called for short: FIFO) doubly linked list etc. the FIFO that metadata can be mapping relations, write operation is corresponding.
Reading field 52, is used for reading to accelerate, and in SSD, accessed frequency (namely reading number of times) is greater than or equal to the page of predetermined threshold value, can be migrated to reading field 52.
Writing region 53, all write operations to SSD, are all first write this to write region 53.Eliminating after completely is write in this region, eliminates by the strategy of nearest minimum accessed frequency.
Clear area 54, this region is by reading field 52 and writes region 53 and shares, and provides idle page for read operation or write operation, and read operation carries out clear area according to reading field 52 with the ratio writing region 53 with write operation and shares.
Supplementary notes, the ratio writing region and reading field in NVM is adjustable.Under normal circumstances, write region much larger than reading field, especially, NVM can only write region, it does not have reading field.It addition, different application scenarios is to performance requirement and the erasable number of times of SSD require it is different.Therefore, based on considering on the service life of SSD, the performance of SSD and cost, can be equipped with according to the practical business demand of user and meet the SSD requiring erasable number of times, by allocating the blend proportion of NVM and NAND, provide the user the product that cost performance is higher, avoid excessively configuring the extra expenses brought, improve resource utilization.
Specifically, controller 42 is for performing the technical scheme that following methods embodiment is protected, for instance, technical scheme as shown in Figure 6, write region with what realize writing data into NVM43, this writes region for storing pre-write data.Wherein, NVM possess by byte (Byte) address, by data with position (bit) for unit write and can step-by-step rewrite ability.
As shown in Figure 6, the embodiment of the present invention provides a kind of method writing data into SSD, and the method can be performed by the controller in SSD.The method includes:
S601, reception write operation requests, this write operation requests includes pre-write data.
S602, determine that these pre-write data are not stored in writing region.
S603, pre-write data are write region.
Specifically, all write operations to SSD40, are all first write region.
For writing the above-mentioned pre-write data of SSD40, these pre-write data are likely to be stored in be write in region, therefore, performs S602, it is first determined these pre-write data are not stored in writing region.
The embodiment of the present invention writes region by pre-write data write NVM's, it is possible to reduce NANDFlash is accessed, and namely reduces the accessed frequency of NANDFlash in SSD, thus reducing the abrasion to NANDFlash;And, due to the non-volatile characteristic of NVM, the data power down being stored in NVM is not lost, and therefore, when power failure, SSD provides standby electricity without electric capacity.For these reasons, increase the service life of embodiment of the present invention SSD.
In the above-described embodiments, the method writing data into SSD can also include: judges whether pre-write data have been stored in NANDFlash;If pre-write data have been stored in NANDFlash, it is invalid to be set to by the page storing pre-write data in NANDFlash.This embodiment can avoid preserving data two parts identical in SSD, in turn results in the waste of memory space.
Further, after S603, the method can also include: preserves mapping relations, and these mapping relations include the logical address of pre-write data and pre-write data write the corresponding relation between the position in region.With reference to Fig. 4, according to these mapping relations, NVM43 and NANDFlash44 can be written and read by controller 42 easily.
Owing to the reading field of NVM43 is used for the data that store accessed frequency (namely reading number of times) greater than or equal to predetermined threshold value, NANDFlash44 is used for the data storing accessed frequency (namely reading number of times) lower than predetermined threshold value, therefore, said method can also include: judges whether the accessed frequency of pre-write data reaches predetermined threshold value;When the accessed frequency of pre-write data reaches predetermined threshold value, migrate pre-write data to reading field, to realize pre-write data from writing zone migration to reading field, accelerate the reading to pre-write data.
In like manner, the method can also include: judges that in NANDFlash, whether the accessed frequency of page is more than or equal to predetermined threshold value;When in NANDFlash, the accessed frequency of page is more than or equal to predetermined threshold value, migrate the accessed frequency reading field more than or equal to the data in the page of predetermined threshold value to NVM, reading field is migrated to realizing data from NANDFlash, accelerate the reading to these data, utilize temporal locality and the spatial locality of data, the data being frequently modified and other hot spot data are stored in NVM, reduce the abrasion to NANDFlash.
It addition, controller 42 can be also used for the step performed as shown in Figure 7.As it is shown in fig. 7, the embodiment of the present invention provides a kind of method writing data into SSD, the method includes:
S701, reception write operation requests, this write operation requests includes pre-write data.
S702, determine that pre-write data have been stored in reading field.
S703, by the pre-write Data Migration in reading field to writing region.
Specifically, it is possible to node corresponding for pre-write data in the second doubly linked list is updated to the afterbody of the first doubly linked list.Wherein, in the first doubly linked list, each node writes the logical address of each page in region for storing;In second doubly linked list, each node is for storing the logical address of each page in reading field.To realize reading field and the mutual migration writing the stored data in region in NVM.
If pre-write data have been stored in the reading field of NVM, are then migrated to and write region, only migrated the node in reading field in implementing to writing region.
The embodiment of the present invention will be stored in the pre-write Data Migration of reading field to writing region, this migration only relates to two zoness of different in NVM, unrelated with NANDFlash, therefore, can reduce and NANDFlash is accessed, namely the accessed frequency of NANDFlash in SSD is reduced, thus reducing the abrasion to NANDFlash;And, due to the non-volatile characteristic of NVM, the data power down being stored in NVM is not lost, and therefore, when power failure, SSD provides standby electricity without electric capacity.For these reasons, increase the service life of embodiment of the present invention SSD.
In above-described embodiment, after S703, the method can also include: updating the first mapping relations is the second mapping relations.Wherein, the first mapping relations include logical address and the pre-write data corresponding relation between the position of reading field of pre-write data;Second mapping relations include logical address and the pre-write data corresponding relation between the position writing region of pre-write data.With reference to Fig. 4, according to these mapping relations, NVM43 and NANDFlash44 can be written and read by controller 42 easily.
Owing to the reading field of NVM43 is used for the data that store accessed frequency (namely reading number of times) greater than or equal to predetermined threshold value, NANDFlash44 is used for the data storing accessed frequency (namely reading number of times) lower than predetermined threshold value, therefore, said method can also include: judges whether the accessed frequency of pre-write data reaches predetermined threshold value;When the accessed frequency of pre-write data reaches predetermined threshold value, migrate pre-write data to reading field, to realize pre-write data from writing zone migration to reading field, accelerate the reading to pre-write Data Data.
In like manner, the method can also include: judges that in NANDFlash, whether the accessed frequency of page is more than or equal to predetermined threshold value;When in NANDFlash, the accessed frequency of page is more than or equal to predetermined threshold value, migrate the accessed frequency reading field more than or equal to the data in the page of predetermined threshold value to NVM, reading field is migrated to realizing data from NANDFlash, accelerate the reading to these data, utilize temporal locality and the spatial locality of data, the data being frequently modified and other hot spot data are stored in NVM, reduce the abrasion to NANDFlash.
Further, NVM43 can also include clear area, and after S703, the method can also include: in clear area, adjusts N page to reading field from keeping for the page writing region, and wherein, N is the number of pages that pre-write data are corresponding.
When having stored above-mentioned pre-write data in SSD, controller 42 also needs to obtain storage position, to determine subsequent operation.Specifically, respectively from following several application scenarios explanations.
In the first application scenarios, pre-write data have been stored in the NANDFlash in SSD.This application scenarios at least includes two kinds of implementations.
In the first implementation, pre-write data take complete page in NANDFlash.Now, update mapping relations, point to pre-write data by the logical address of above-mentioned pre-write data and write the position in region.Wherein, mapping relations include the corresponding relation of the logical address of all pages and physical address in SSD.
In the second implementation, pre-write data take incomplete page in NANDFlash.Now, Data Migration except the part identical with the scope of the logical address of pre-write data in page corresponding with pre-write data in NANDFlash is write region to NVM;In writing region, combine written into pre-write data and migrating to write the page that the data composition in region is new;Then, the sensing of the logical address of pre-write data in mapping relations is updated so that it is point to the address of above-mentioned new page.
Wherein, above two implementation is distinctive in that: the first implementation is for illustrating the integral multiple that the size of pre-write data is exactly in NANDFlash page, now, only need to update the corresponding relation of logical address and physical address in mapping relations, in mapping relations, increase the logical address of these pre-write data and the corresponding relation of the physical address in NVM thereof;The second implementation is for illustrating that the size of pre-write data is not the process during integral multiple of page in NANDFlash.
Such as, pre-write data be sized to 4KB, and these pre-write data have been stored in page 1 in NANDFlash, in NANDFlash page be sized to 8KB, page 1 is a corresponding logical address and physical address in mapping relations.When writing these pre-write data, by Data Migration except the part identical with the scope of the logical address of pre-write data in page 1 to writing region;In writing region, the page that combination pre-write data are new with migrating to the data composition writing region;Then, update mapping relations and preset the sensing of the logical address of pre-write data in mapping relations so that it is the address of page new after pointing to combination.
Alternatively, the address of above-mentioned new page being preserved the afterbody to the first doubly linked list, in this first doubly linked list, each node writes the logical address of each page in region for storing.
Specifically, in the first doubly linked list, each node writes one page in region for expression.Node has predecessor pointers and heir pointer, and predecessor pointers points to forerunner's node of the first doubly linked list, and heir pointer points to the successor node of the first doubly linked list;It addition, node also includes the logical address of page.The size writing region determines the number of the node comprised in the first doubly linked list.The first doubly linked list record is adopted to write the sequencing of each page in region, thus convenient to the management writing region.Such as, when write data, the information of corresponding page can be inserted the afterbody of the first doubly linked list;When eliminating, page can be chosen from the stem of the first doubly linked list.
In the second application scenarios, the NVM that pre-write data have been stored in SSD writes region.Under this application scenarios, at least include implementations below: receiving the write operation requests that requesting party sends, this write operation requests includes pre-write data;Determine that above-mentioned pre-write data have been stored in and write region;The position storing pre-write data in writing region re-writes pre-write data.
Afterbody by knot refinement corresponding for pre-write data in the first doubly linked list to the first doubly linked list.In this application scenarios, it is achieved write the internal migration of the stored data in region.If pre-write data have been stored in and have write region, then the node that it is corresponding in the first doubly linked list is migrated to the afterbody of the first doubly linked list.
Alternatively, after S603, the method can also include: the address of page corresponding for pre-write data is preserved the afterbody to the first doubly linked list.
And, according to the number of pages that pre-write data are corresponding, update the counting written into page in NVM;If judging, counting is more than or equal to first threshold, then from the page of NVM migration first threshold number to NANDFlash, there is enough space for the migration of the stored data in other regions in the write of new data and/or SSD realizing NVM, it is achieved SSD stores dynamically adjusting of data.
Especially, before pre-write data write region, if it is full to judge to write region, from the stem of the first doubly linked list, judge that the reading number of times of the page that in the first doubly linked list, node is corresponding is more than or equal to Second Threshold, then migrate data in the page of number of pages corresponding to pre-write data to reading field;Preserving the address of page corresponding in reading field for the data in page to the second doubly linked list, in this second doubly linked list, each node is for storing the logical address of each page in reading field.Similar with writing region, corresponding second doubly linked list of reading field, the structure of this second doubly linked list is identical with the structure of the first doubly linked list, repeats no more herein.
Migrate data in the page of number of pages corresponding to pre-write data to before reading field, if judging, reading field is full, or, if it is full to judge to write region, from the stem of the first doubly linked list, judge that the reading number of times of the page that in the first doubly linked list, node is corresponding is less than Second Threshold, then from the stem of the second doubly linked list, migrate data in the page of number of pages corresponding to pre-write data to NANDFlash, with data in real-time update reading field, when memory space inadequate, in time by Data Migration low for accessed frequency to NANDFlash, it is further ensured that the performance of SSD.
Meanwhile, cumulative corresponding for NANDFlash migration number of times;If judging to migrate number of times more than or equal to the 3rd threshold value, then choose the page migration of the 3rd threshold number from NVM to NANDFlash, the selected page that page is the 3rd threshold number started from the stem writing the first doubly linked list corresponding to region, wherein, the 3rd threshold value ratio-dependent shared by NVM and NANDFlash in SSD.This embodiment can balance the stored data of NANDFlash and NVM in SSD, it is to avoid data are stored entirely in NVM, causes the waste of the memory space of NANDFlash.
In this second application scenarios, pre-write data have been stored in and have write region, now, while covering is write, consider that in each region and NANDFlash, whether memory space is sufficient in NVM, and carry out the adjustment of wherein stored data according to the above description, thus ensureing that in SSD, each storage medium has enough spaces to meet the demand.
To sum up, it is stored in the data in SSD, is substantially stored in NVM or NANDFlash.Specifically, NVM preserves the data that accessed frequency is high, and NANDFlash preserves the data that accessed frequency is relatively low.The height of accessed frequency can judge in several ways, for instance, set a predetermined threshold value, contrast according to the accessed frequency of data in the set time with this predetermined threshold value, obtaining the storage position of data, now, the accessed frequency of data place page is saved among above-mentioned mapping relations;Or, the accessed frequency of data place page position response data in above-mentioned mapping relations, etc., the present invention is restriction not.Below with the reading to having stored data in SSD of the present invention of several specific embodiments.
On the basis of above-described embodiment, the method can also include: the reading number of times of the page at cumulative pre-reading data place, wherein, above-mentioned mapping relations include the reading number of times of all pages in SSD.In this embodiment, the accessed frequency of data can be passed through to read number of times and embody, and this reading number of times is included in default mapping relations.Alternatively, the accessed frequency of data can also be realized by other form, repeats no more herein.
Wherein, mapping relations can be expressed as the form of mapping table as shown in Figure 8.In fig. 8, each square frame represents a basic map unit, and its size is such as 4KB or 8KB.Accordingly, it is also possible to use a structure to represent the field corresponding with default mapping relations.Wherein, the field that single mapping relations are corresponding includes the physical address of NVM or NAND;For indicating the page that this square frame is corresponding to be positioned at the indicating bit of NVM or NAND, for instance, this indicating bit can represent with bit (bit) position;And the reading counting of page corresponding to this square frame.It should be noted that, the size of page changes with the change of the size of page in NAND.In prior art page be typically sized to 4KB/8KB, but the present invention is not limited system.
In a kind of embodiment, the physical address of pre-reading data is positioned at NANDFlash.After the reading number of times of the page at cumulative pre-reading data place, the method can also include: the every one page to pre-reading data place, if judging, the reading number of times of the page at pre-reading data place is more than or equal to Second Threshold, and available free page in NVM, then will read number of times more than or equal to free page in the Data Migration in the page of Second Threshold to NVM.Former that be stored in NANDFlash, accessed frequency is higher than the Data Migration of Second Threshold to NVM by the embodiment of the present invention, to realize storing in SSD the dynamic adjustment of data, it is to avoid NANDFlash is too much accessed.The number of times data more than or equal to Second Threshold will be read, preserve to NVM, relative to being stored in NANDFlash, the speed of digital independent can be improved, utilize temporal locality and the spatial locality of data, the data being frequently modified and other hot spot data are stored in NVM, reduce the abrasion to NANDFlash.
Alternatively, by after Data Migration in NANDFlash to NVM, it is also possible to cumulative corresponding for NANDFlash migration number of times.Such that the number of pages migrating to NVM from NANDFlash can be obtained, thus the ratio in space shared by NANDFlash and NVM in Reasonable adjustment SSD, or, by the Data Migration of number of pages equal in NVM to NANDFlash.
Further, when NVM includes reading field, if it is determined that the reading number of times of the page at pre-reading data place is more than or equal to Second Threshold, and available free page in NVM, then will read number of times more than or equal to the Data Migration in the page of Second Threshold to NVM, if may include that, the reading number of times of the page judging pre-reading data place is more than or equal to Second Threshold, and reading field includes free page, then will read the number of times free page more than or equal to the Data Migration in the page of Second Threshold to reading field.
On the basis of the above, however, it is determined that above-mentioned migration number of times more than or equal to the 3rd threshold value, then chooses the page migration of the 3rd threshold number to NANDFlash from NVM, wherein, the 3rd threshold value ratio-dependent shared by NVM and NANDFlash in SSD.In SSD, if the space that NVM takies is more than the NANDFlash space taken, the 3rd threshold value is less;Accordingly, if the space that NVM takies is less than the NANDFlash space taken, the 3rd threshold value is bigger.
In another kind of embodiment, NVM includes writing region.If what the physical address of pre-reading data was positioned at NVM writes region, after the reading number of times of the page at above-mentioned cumulative pre-reading data place, the method can also include: updates the address of page at pre-reading data place to the afterbody of the first doubly linked list, wherein, in the first doubly linked list, each node writes the logical address of each page in region for storing.
By above-described embodiment it can be seen that metadata area can store above-mentioned first doubly linked list, the second doubly linked list and default mapping relations etc..
Summary embodiment, the embodiment of the present invention stores the data that accessed frequency is high in NVM, it is thus possible to reduce, NANDFlash is accessed, and namely reduces the accessed frequency of NANDFlash in SSD, reduces the abrasion to NANDFlash;And, due to the non-volatile characteristic of NVM, the data power down being stored in NVM is not lost, and therefore, when power failure, SSD provides standby electricity without electric capacity.For these reasons, increase the service life of embodiment of the present invention SSD.
In actual applications, the embodiment of the present invention size to reading and write is not do any restriction, it is possible to be reading or the write of any size.But, there is a basic management granularity, such as 4KB, or 8KB, even 16KB in the default mapping relations of SSD, for instance mapping table.This management granularity generally size with the page (page) of NANDFlash is the same, it is also possible to different.It addition, the basic management unit of SDD read-write is page.
The size of the read-write operation request that main frame is issued on SSD is probably arbitrarily.Illustrate for the 8KB that is sized to of page: if the data to read are relatively larger, for 32KB, then need to read 4 pages, and relevant data are gathered together feed back to main frame;If the data read are smaller, for 2KB, then only the data of this corresponding 2KB are returned to main frame.
In the embodiment of the present invention, SSD adopts NANDFlash and NVM that data are carried out classification storage, namely, data are stored in NANDFlash or NVM, non-volatile characteristic because of NVM, thus when not comprising electric capacity in SSD, without producing the problem that data power down is lost, and SSD does not comprise electric capacity also can increase the service life of SSD, the reliability of significant increase SSD.
In several embodiments provided herein, it should be understood that disclosed equipment and method, it is possible to realize by another way.Such as, apparatus embodiments described above is merely schematic, such as, the division of described unit or module, being only a kind of logic function to divide, actual can have other dividing mode when realizing, for instance multiple unit or module can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some interfaces, equipment or module, it is possible to be electrical, machinery or other form.
The described module illustrated as separating component can be or may not be physically separate, and the parts shown as module can be or may not be physical module, namely may be located at a place, or can also be distributed on multiple NE.Some or all of module therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.
One of ordinary skill in the art will appreciate that: all or part of step realizing above-mentioned each embodiment of the method can be completed by the hardware that programmed instruction is relevant.Aforesaid program can be stored in a computer read/write memory medium.This program upon execution, performs to include the step of above-mentioned each embodiment of the method;And aforesaid storage medium includes: the various media that can store program code such as ROM, RAM, magnetic disc or CDs.
Last it is noted that various embodiments above is only in order to illustrate technical scheme, it is not intended to limit;Although the present invention being described in detail with reference to foregoing embodiments, it will be understood by those within the art that: the technical scheme described in foregoing embodiments still can be modified by it, or wherein some or all of technical characteristic is carried out equivalent replacement;And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (20)

1. the method writing data into solid-state hard disk SSD, described SSD includes controller and memorizer, it is characterised in that the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM and includes writing region, and described method includes:
Receiving write operation requests, described write operation requests includes pre-write data;
Determine that described pre-write data write region described in being not stored in;
Region is write described in described pre-write data being write.
2. method according to claim 1, it is characterised in that described method also includes:
Judge whether described pre-write data have been stored in described NANDFlash;
If described pre-write data have been stored in described NANDFlash, it is invalid to be set to by the page storing described pre-write data in NANDFlash.
3. method according to claim 1 and 2, it is characterised in that described by described pre-write data write described in write region after, described method also includes:
Preserving mapping relations, described mapping relations include the corresponding relation writing between the position in region described in the logical address of described pre-write data and the write of described pre-write data.
4. the method according to any one of claim 1-3, it is characterised in that described method also includes:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
5. the method according to any one of claim 1-4, it is characterised in that described method also includes:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
6. the method writing data into solid-state hard disk SSD, described SSD includes controller and memorizer, it is characterised in that the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM includes writing region and reading field, and described method includes:
Receiving write operation requests, described write operation requests includes pre-write data;
Determine that described pre-write data have been stored in described reading field;
By the described pre-write Data Migration in described reading field to described in write region.
7. method according to claim 6, it is characterised in that described by the described pre-write Data Migration in described reading field to described in write region after, described method also includes:
Updating the first mapping relations is the second mapping relations, described first mapping relations include logical address and the described pre-write data corresponding relation between the position of described reading field of described pre-write data, and described second mapping relations include logical address and the described pre-write data corresponding relation between the described position writing region of described pre-write data.
8. the method according to claim 6 or 7, it is characterised in that described method also includes:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
9. the method according to any one of claim 6-8, it is characterised in that described method also includes:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
10. the method according to any one of claim 6-9, it is characterised in that described NVM also includes clear area, described by the described pre-write Data Migration in described reading field to described in write region after, described method also includes:
In described clear area, adjusting N page to described reading field from the page writing region described in keeping for, described N is the number of pages that described pre-write data are corresponding.
11. a solid-state hard disk SSD, described SSD includes controller and memorizer, it is characterised in that the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, and described NVM includes writing region, wherein,
Described controller, is used for receiving write operation requests, and described write operation requests includes pre-write data;Determine that described pre-write data write region described in being not stored in;And, write region described in described pre-write data being write;
Described region of writing is for storing described pre-write data.
12. SSD according to claim 11, it is characterised in that described controller is additionally operable to:
Judge whether described pre-write data have been stored in described NANDFlash;
If described pre-write data have been stored in described NANDFlash, it is invalid to be set to by the page storing described pre-write data in NANDFlash.
13. the SSD according to claim 11 or 12, it is characterised in that described controller is additionally operable to:
Preserving mapping relations, described mapping relations include the corresponding relation writing between the position in region described in the logical address of pre-write data and the write of described pre-write data.
14. according to the SSD described in any one of claim 11-13, it is characterised in that described controller is additionally operable to:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
15. according to the SSD described in any one of claim 11-14, it is characterised in that described controller is additionally operable to:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
16. a solid-state hard disk SSD, described SSD includes controller and memorizer, it is characterised in that the storage medium of described memorizer includes NANDFlash and nonvolatile memory NVM, described NVM and includes writing region and reading field, wherein,
Described controller, is used for receiving write operation requests, and described write operation requests includes pre-write data;Determine that described pre-write data have been stored in described reading field;And, by the described pre-write Data Migration in described reading field to described in write region.
17. SSD according to claim 16, it is characterised in that described controller is additionally operable to:
Updating the first mapping relations is the second mapping relations, described first mapping relations include logical address and the described pre-write data corresponding relation between the position of described reading field of described pre-write data, and described second mapping relations include logical address and the described pre-write data corresponding relation between the described position writing region of described pre-write data.
18. the SSD according to claim 16 or 17, it is characterised in that described controller is additionally operable to:
Judge whether the accessed frequency of described pre-write data reaches predetermined threshold value;
When the accessed frequency of described pre-write data reaches described predetermined threshold value, migrate described pre-write data extremely described reading field.
19. according to the SSD described in any one of claim 16-18, it is characterised in that described controller is additionally operable to:
Judge that in described NANDFlash, whether the accessed frequency of page is more than or equal to described predetermined threshold value;
When the accessed frequency of page is more than or equal to described predetermined threshold value in described NANDFlash, migrate the described accessed frequency reading field more than or equal to the data in the page of described predetermined threshold value to described NVM.
20. according to the SSD described in any one of claim 16-19, it is characterised in that described NVM also includes clear area, and described controller is additionally operable to:
In described clear area, adjusting N page to described reading field from the page writing region described in keeping for, described N is the number of pages that described pre-write data are corresponding.
CN201410852946.0A 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk Active CN105808156B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201410852946.0A CN105808156B (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk
CN202010311530.3A CN111638852A (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk
PCT/CN2015/098109 WO2016107442A1 (en) 2014-12-31 2015-12-21 Method for writing data to solid state drive and solid state drive

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410852946.0A CN105808156B (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk

Related Child Applications (1)

Application Number Title Priority Date Filing Date
CN202010311530.3A Division CN111638852A (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk

Publications (2)

Publication Number Publication Date
CN105808156A true CN105808156A (en) 2016-07-27
CN105808156B CN105808156B (en) 2020-04-28

Family

ID=56284227

Family Applications (2)

Application Number Title Priority Date Filing Date
CN202010311530.3A Pending CN111638852A (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk
CN201410852946.0A Active CN105808156B (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk

Family Applications Before (1)

Application Number Title Priority Date Filing Date
CN202010311530.3A Pending CN111638852A (en) 2014-12-31 2014-12-31 Method for writing data into solid state disk and solid state disk

Country Status (2)

Country Link
CN (2) CN111638852A (en)
WO (1) WO2016107442A1 (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109119106A (en) * 2017-06-26 2019-01-01 中兴通讯股份有限公司 Date storage method, equipment and computer readable storage medium
CN109144422A (en) * 2018-08-31 2019-01-04 浪潮电子信息产业股份有限公司 Data writing method, device and equipment for solid state disk and storage medium
CN110109915A (en) * 2018-01-18 2019-08-09 伊姆西Ip控股有限责任公司 For managing the method, equipment and computer program product of Hash table
CN110515549A (en) * 2019-08-16 2019-11-29 浪潮电子信息产业股份有限公司 A kind of data migration method and device
CN110908926A (en) * 2018-09-14 2020-03-24 慧荣科技股份有限公司 Data storage device and writing method of logical-to-physical address mapping table
CN110989924A (en) * 2019-10-30 2020-04-10 烽火通信科技股份有限公司 Metadata storage performance optimization method and storage server
CN111208940A (en) * 2019-12-24 2020-05-29 华东师范大学 Read-write separation method based on multi-channel flash memory
CN111367463A (en) * 2018-12-25 2020-07-03 北京兆易创新科技股份有限公司 Storage space management method and device
CN111459845A (en) * 2019-01-22 2020-07-28 爱思开海力士有限公司 Storage device, computing system including the same, and operating method thereof
CN111897744A (en) * 2019-05-05 2020-11-06 北京兆易创新科技股份有限公司 Data writing method and device, electronic equipment and storage medium
CN112717419A (en) * 2021-01-04 2021-04-30 厦门梦加网络科技股份有限公司 Game data storage method
CN113254364A (en) * 2021-05-24 2021-08-13 山东创恒科技发展有限公司 Information storage device for embedded system
WO2021232427A1 (en) * 2020-05-22 2021-11-25 Yangtze Memory Technologies Co., Ltd. Flush method for mapping table of ssd
WO2022021280A1 (en) * 2020-07-31 2022-02-03 华为技术有限公司 Storage controller, storage control method, solid state disk and storage system
CN114328281A (en) * 2021-11-30 2022-04-12 苏州浪潮智能科技有限公司 Method and device for processing abnormal power failure of solid state disk, electronic equipment and medium
CN115904255A (en) * 2023-01-19 2023-04-04 苏州浪潮智能科技有限公司 Data request method, device, equipment and storage medium
CN116627354A (en) * 2023-07-20 2023-08-22 湖南博匠信息科技有限公司 High-performance data writing method and data recording equipment for solid state disk

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI607448B (en) * 2016-09-07 2017-12-01 群聯電子股份有限公司 Data writing method, memory control circuit unit and mempry storage apparatus
CN110286995B (en) * 2018-03-19 2022-02-25 北京兆易创新科技股份有限公司 Data reading and writing method and device
CN110389712B (en) * 2018-04-20 2022-12-23 杭州海康存储科技有限公司 Data writing method and device, solid state disk and computer readable storage medium
CN108984124B (en) * 2018-07-13 2022-03-25 深圳忆联信息***有限公司 Method and device for reducing write amplification, computer equipment and storage medium
CN111581117A (en) * 2019-02-19 2020-08-25 睿宽智能科技有限公司 Solid state disk unmapped address redirection method
CN110795045B (en) * 2019-10-30 2024-04-09 中国科学院微电子研究所 Data migration method and system of hybrid memory and electronic equipment
CN111143279B (en) * 2019-12-29 2022-04-22 浪潮电子信息产业股份有限公司 Data migration method, device and equipment and readable storage medium
CN112435706B (en) * 2020-11-05 2023-06-30 北京中电华大电子设计有限责任公司 Nonvolatile memory application level erasing life evaluation method
CN113760782A (en) * 2021-08-23 2021-12-07 南京森根科技股份有限公司 Dynamically adjustable annular cache system and control method thereof
CN113835639B (en) * 2021-09-26 2024-03-19 深圳大普微电子科技有限公司 I/O request processing method, device, equipment and readable storage medium
CN115657972B (en) * 2022-12-27 2023-06-06 北京特纳飞电子技术有限公司 Solid state disk writing control method and device and solid state disk

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101673188A (en) * 2008-09-09 2010-03-17 上海华虹Nec电子有限公司 Data access method for solid state disk
CN102360273A (en) * 2011-09-26 2012-02-22 北京邦诺存储科技有限公司 Equipment and method for controlling cooperative storage of memory and nonvolatile caching device
CN102646069A (en) * 2012-02-23 2012-08-22 华中科技大学 Method for prolonging service life of solid-state disk
US20120254523A1 (en) * 2008-02-27 2012-10-04 Kentaro Shimada Storage system which utilizes two kinds of memory devices as its cache memory and method of controlling the storage system
CN103377152A (en) * 2012-04-26 2013-10-30 深圳市朗科科技股份有限公司 Write operation control method and write operation device for solid state disk
CN103885728A (en) * 2014-04-04 2014-06-25 华中科技大学 Magnetic disk cache system based on solid-state disk
US8856423B1 (en) * 2007-04-10 2014-10-07 Marvell International Ltd. Dual-purpose nonvolatile memory for code and data storage

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20100007409A (en) * 2008-07-14 2010-01-22 엘지전자 주식회사 Storage apparatus and method of media file
KR101515525B1 (en) * 2008-10-02 2015-04-28 삼성전자주식회사 Memory device and operating method of memory device
CN102955787A (en) * 2011-08-22 2013-03-06 陈思达 Using method for file directory table, file writing method and main circuit board, CPU (central processing unit) and external storage applying same
CN102591748A (en) * 2011-12-29 2012-07-18 记忆科技(深圳)有限公司 Solid state disc and power failure protection method and system thereof
CN103473186A (en) * 2012-06-07 2013-12-25 鸿富锦精密工业(深圳)有限公司 SSD (solid state disc) data protection circuit
CN102981969A (en) * 2012-11-21 2013-03-20 记忆科技(深圳)有限公司 Method for deleting repeated data and solid hard disc thereof

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8856423B1 (en) * 2007-04-10 2014-10-07 Marvell International Ltd. Dual-purpose nonvolatile memory for code and data storage
US20120254523A1 (en) * 2008-02-27 2012-10-04 Kentaro Shimada Storage system which utilizes two kinds of memory devices as its cache memory and method of controlling the storage system
CN101673188A (en) * 2008-09-09 2010-03-17 上海华虹Nec电子有限公司 Data access method for solid state disk
CN102360273A (en) * 2011-09-26 2012-02-22 北京邦诺存储科技有限公司 Equipment and method for controlling cooperative storage of memory and nonvolatile caching device
CN102646069A (en) * 2012-02-23 2012-08-22 华中科技大学 Method for prolonging service life of solid-state disk
CN103377152A (en) * 2012-04-26 2013-10-30 深圳市朗科科技股份有限公司 Write operation control method and write operation device for solid state disk
CN103885728A (en) * 2014-04-04 2014-06-25 华中科技大学 Magnetic disk cache system based on solid-state disk

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109119106B (en) * 2017-06-26 2022-08-02 中兴通讯股份有限公司 Data storage method, device and computer readable storage medium
CN109119106A (en) * 2017-06-26 2019-01-01 中兴通讯股份有限公司 Date storage method, equipment and computer readable storage medium
CN110109915B (en) * 2018-01-18 2024-01-05 伊姆西Ip控股有限责任公司 Method, apparatus and computer program product for managing hash tables
CN110109915A (en) * 2018-01-18 2019-08-09 伊姆西Ip控股有限责任公司 For managing the method, equipment and computer program product of Hash table
CN109144422A (en) * 2018-08-31 2019-01-04 浪潮电子信息产业股份有限公司 Data writing method, device and equipment for solid state disk and storage medium
CN110908926A (en) * 2018-09-14 2020-03-24 慧荣科技股份有限公司 Data storage device and writing method of logical-to-physical address mapping table
CN111367463B (en) * 2018-12-25 2023-05-09 兆易创新科技集团股份有限公司 Storage space management method and device
CN111367463A (en) * 2018-12-25 2020-07-03 北京兆易创新科技股份有限公司 Storage space management method and device
CN111459845A (en) * 2019-01-22 2020-07-28 爱思开海力士有限公司 Storage device, computing system including the same, and operating method thereof
CN111897744A (en) * 2019-05-05 2020-11-06 北京兆易创新科技股份有限公司 Data writing method and device, electronic equipment and storage medium
CN110515549A (en) * 2019-08-16 2019-11-29 浪潮电子信息产业股份有限公司 A kind of data migration method and device
CN110515549B (en) * 2019-08-16 2022-03-22 浪潮电子信息产业股份有限公司 Data migration method and device
CN110989924A (en) * 2019-10-30 2020-04-10 烽火通信科技股份有限公司 Metadata storage performance optimization method and storage server
CN110989924B (en) * 2019-10-30 2022-10-18 烽火通信科技股份有限公司 Metadata storage performance optimization method and storage server
CN111208940A (en) * 2019-12-24 2020-05-29 华东师范大学 Read-write separation method based on multi-channel flash memory
WO2021232427A1 (en) * 2020-05-22 2021-11-25 Yangtze Memory Technologies Co., Ltd. Flush method for mapping table of ssd
WO2022021280A1 (en) * 2020-07-31 2022-02-03 华为技术有限公司 Storage controller, storage control method, solid state disk and storage system
EP4187363A4 (en) * 2020-07-31 2023-09-06 Huawei Technologies Co., Ltd. Storage controller, storage control method, solid state disk and storage system
CN112717419A (en) * 2021-01-04 2021-04-30 厦门梦加网络科技股份有限公司 Game data storage method
CN113254364A (en) * 2021-05-24 2021-08-13 山东创恒科技发展有限公司 Information storage device for embedded system
CN114328281A (en) * 2021-11-30 2022-04-12 苏州浪潮智能科技有限公司 Method and device for processing abnormal power failure of solid state disk, electronic equipment and medium
CN114328281B (en) * 2021-11-30 2023-11-14 苏州浪潮智能科技有限公司 Solid state disk abnormal power failure processing method and device, electronic equipment and medium
CN115904255A (en) * 2023-01-19 2023-04-04 苏州浪潮智能科技有限公司 Data request method, device, equipment and storage medium
CN116627354A (en) * 2023-07-20 2023-08-22 湖南博匠信息科技有限公司 High-performance data writing method and data recording equipment for solid state disk
CN116627354B (en) * 2023-07-20 2023-09-15 湖南博匠信息科技有限公司 High-performance data writing method and data recording equipment for solid state disk

Also Published As

Publication number Publication date
WO2016107442A1 (en) 2016-07-07
CN105808156B (en) 2020-04-28
CN111638852A (en) 2020-09-08

Similar Documents

Publication Publication Date Title
CN105808156A (en) Method for writing data into solid state drive and solid state drive
CN107844431B (en) Mapping table updating method, memory control circuit unit and memory storage device
US11042306B2 (en) Memory management
US8606987B2 (en) Data writing method for flash memory and controller using the same
US8463826B2 (en) Incremental garbage collection for non-volatile memories
US8180953B2 (en) Data accessing method for flash memory, and storage system and controller system thereof
US8650379B2 (en) Data processing method for nonvolatile memory system
TW201732597A (en) Data storage device and operating method thereof
US20230176965A1 (en) Media management based on data access metrics
CN111158579B (en) Solid state disk and data access method thereof
US9619165B1 (en) Convertible leaf memory mapping
CN111309654B (en) Memory device and method of operating the same
CN105808455A (en) Memory access method, storage-class memory and computer system
US8423707B2 (en) Data access method for flash memory and storage system and controller using the same
US11537305B1 (en) Dissimilar write prioritization in ZNS devices
CN103365786A (en) Data storage method, device and system
CN105278875A (en) Hybrid heterogeneous NAND solid state device
US11507289B2 (en) Storage device, controller and method for operating storage device
US11537303B1 (en) Dissimilar write prioritization in ZNS devices
CN113590505B (en) Address mapping method, solid state disk controller and solid state disk
US11829650B2 (en) Memory sub-system data migration
US11036414B2 (en) Data storage device and control method for non-volatile memory with high-efficiency garbage collection
US8713242B2 (en) Control method and allocation structure for flash memory device
US20240062823A1 (en) Mixed write cursor for block stripe writing
US20240061589A1 (en) Code rate as function of logical saturation

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant