CN106527971B - A method of promoting SSD reading performance during garbage reclamation - Google Patents
A method of promoting SSD reading performance during garbage reclamation Download PDFInfo
- Publication number
- CN106527971B CN106527971B CN201610877009.XA CN201610877009A CN106527971B CN 106527971 B CN106527971 B CN 106527971B CN 201610877009 A CN201610877009 A CN 201610877009A CN 106527971 B CN106527971 B CN 106527971B
- Authority
- CN
- China
- Prior art keywords
- data
- page
- read request
- raid
- read
- 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.)
- Active
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
- G06F3/0683—Plurality of storage devices
- G06F3/0689—Disk arrays, e.g. RAID, JBOD
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
The invention discloses a kind of methods for promoting SSD reading performance during garbage reclamation, comprising: establishes RAID by band of data page in each interchannel；When handling read request, if the wafer (die) where the data page of read request access is not carrying out erasing operation, read request normal response at this time；If the wafer (die) where the data page of read request access is just being wiped free of occupied by operation, short duration failure occurs, RAID reconstruction is carried out by reading the data of other data pages on same band using the RAID of interchannel at this time, to timely respond to the read request.Method of the invention reduces Time Delay of Systems caused by garbage collection operation in the solid storage device based on NAND Flash, solves the collision problem of system normal read request and garbage reclamation erasing operation using the RAID technique in SSD.
Description
Technical field
The invention belongs to solid-state disk (SSD) readwrite performance control technology fields, and in particular to a kind of promotion SSD is returned in rubbish
The method of reading performance during receipts.
Background technique
In recent years, the solid state hard disk (Solid State Drive, SSD) based on NAND Flash is since performance is high, function
Many advantages, such as consuming good low, high reliablity, shock resistance, noiseless is widely applied.But SSD is using a period of time
Afterwards, access performance can be decreased significantly.Garbage reclamation has significant impact to the performance of SSD, by Intel SSD DC
P3700 (800G) carries out test discovery, is 4KB for read-write granularity, and read-write ratio is the random mixing request of 7:3.Work as Intel
When SSD is blank panel, IOPS can reach 200,000, and readwrite bandwidth can reach 800MB/s；It is pre-buried when carrying out data to Intel SSD
After fragmentation processing, IOPS drops to 9W, and readwrite bandwidth drops to 360MB/s.Main cause is to use a period of time
Afterwards, the frequent triggering of the inside SSD garbage reclamation, reduces the readwrite performance of SSD.
Due to generalling use strange land more new strategy when flash chip, which has, writes the preceding characteristic, more new data wiped, this can be produced
Raw a large amount of invalid pages.For the normal use for guaranteeing SSD, needs to carry out garbage collection operation to invalid page, wipe the target of selection
Block is to reuse.Since the read-write granularity of flash chip is different with erasing granularity, before wiping a block, garbage reclamation one
As be related to the migration of valid data, in Data Migration, chip locating for the source Physical Page and purpose Physical Page of data
(chip) and channel (channel) can be occupied.When executing erasing operation, although will not active channel resource, erasing pair
As locating wafer (die) can be occupied.Therefore, the readwrite performance of SSD will be a greater impact in garbage collection procedure.
As shown in figure 3, a complete garbage collection operation includes three steps: 1) the selection object block to be recycled, 2) it carries out
The migration of valid data, 3) erasing object block.When wiping object block, entire die will be occupied, be easy to cause request collision,
It cannot timely respond to extraneous request.The conflict of write request can be solved by modes such as caching and reserved spaces, be read in comparison
Conflict more difficult, can only generally be forced that erasing operation is waited to complete, in addition synchronous read operation not only influences this channel
(channel) request also will affect the response of each channel (channel).When comparing the erasing operation and read operation of Flash
Between, the erasing operation time is ms grade, and read time is us grades, and the two time ratio is close to 100:1；And with SSD capacity
Increase, erasing block size is further enlarged, and the time phase difference of the two can be increasing.In conclusion being forced for read request
Wait the completion of erasing operation that can significantly reduce the access performance of SSD.
Because garbage reclamation is related to the migration and block erasing operation of valid data, a large amount of system resources are occupied, to the greatest extent may be used
Can be reduced influence of the garbage reclamation to SSD access performance, presently, there are make full use of system idling-resource carry out garbage reclamation or
By utmostly improving the scheme of the garbage reclamation of garbage reclamation efficiency, but garbage reclamation is considered as always in these schemes
One can not be split entirety, lack the considerations of garbage reclamation conflicts to SSD normal request.To solve the above problems, the prior art
In there is a kind of interruptable garbage reclamation mechanism, during valid data migration, one data page of every migration can in
Ongoing garbage reclamation of breaking responds normal read-write requests.The program can solve garbage reclamation and just to a certain extent
The conflict often requested, but it, which interrupts selection of time, limitation, can only occur in Data Migration, not account for garbage reclamation
The conflict of time-consuming longest erasing operation and read-write requests in the process.
Summary of the invention
Aiming at the above defects or improvement requirements of the prior art, the present invention provides a kind of promotion SSD in the garbage reclamation phase
Between the method for reading performance, so as to timely respond to read request, solved current by building RAID in solid-state disk interchannel
SSD during garbage reclamation read request performance it is low, overcome in garbage collection procedure existing in the prior art time-consuming longest
The conflict defect of erasing operation and read-write requests.
To achieve the above object, it is proposed, according to the invention, a kind of method for promoting SSD reading performance during garbage reclamation is provided,
Include:
In single solid-state disk SSD inside configuration, each interchannel of flash controller control is established by band of data page
RAID；
When handling read request, if the wafer (die) where the data page of read request access does not carry out erasing operation, this
When read request normal response；
If the wafer (die) where the data page of read request access is just being wiped free of occupied by operation, short duration failure occurs,
RAID reconstruction is carried out by reading the data of other data pages on same band using the RAID of interchannel at this time, thus and
When respond the read request.
As present invention further optimization, the RAID, which is rebuild, passes through other data page exclusive or in RAID stripe
Mode restore the data page read required for the read request.
As present invention further optimization, if erasing operation is completed in carrying out RAID reconstruction process, wafer (die) can
Again it is accessed, then eliminates the RAID and rebuild trace, normal response read request.
As present invention further optimization, the RAID in each channel rebuild read operation can parallel processing, and do not influence
Normal read-write requests on the channel, sequencing can be had by rebuilding read operation completion.
As present invention further optimization, verification data page data is calculated or updated in write request per treatment,
The calculating or update verification data page data include that entire band writes data for the first time, individual data page writes number for the first time
According to, individual data web update data and entire band more new data.
As present invention further optimization, the entire band first time data packets include while writing number to each data page
According to verification page data does XOR operation by each data page data and obtains, and each data page and verification page are finally constituted a band
Flash memory is written.
As present invention further optimization, the individual data page first time data packets include one into any bar band
Data are written in a new data page, and original checksums page data is read DRAM from flash memory first, then verify the check number of page
According to obtaining newly verifying data with the new data page data exclusive or, finally by the new data page data and new verification data write-in
Flash memory.
As present invention further optimization, the individual data web update data packets are included the individual data page
Former data and the former verification data of verification page are read out from flash memory, then by the former data and new data step-by-step exclusive or,
Result after exclusive or is again with original verification data step-by-step exclusive or, and obtained result is new verification data, finally by new data and newly
Verification data write back in flash memory.
As present invention further optimization, the entire band is updated data packets and includes while being updated using new data each
Data page, verification page data are obtained by the new data exclusive or of each data web update, finally by the new data of each data web update and
New verification page data writes back flash memory.
It is another aspect of this invention to provide that a kind of method by the built-in RAID response system read request of interchannel is provided,
Its step are as follows:
1) in single solid-state disk SSD inside configuration, each interchannel of flash controller control is built by band of data page
Vertical RAID；
2) there is read request, set counter i=1；
3) whether the wafer (die) where judging the read request, which is wiped free of operation, occupies, and is to go to step 4), and no step turns
9)；
4) judge whether i is equal to 1, be, go to step 5), otherwise go to step 6)；
5) it generates n-1 reconstruction read request to be mounted to respectively on n-1 channel, reads the data of remaining RAID stripe,
Middle n is total number of channels；
6) n-i channel of parallel processing, successively existing read-write requests in treatment channel；
7) respective data blocks of same band on channel, the reconstruction read request on one channel of every processing, i++ are read；
8) judge whether n-i is equal to 0, be, go to step 10), otherwise go to step 3)；
9) data block is directly read, the trace that RAID is rebuild is removed, goes to step 11)；
10) according to existing n-1 data block, restore the data block to be read using RAID5；
11) read request is responded.
In general, through the invention it is contemplated above technical scheme is compared with the prior art, have below beneficial to effect
Fruit:
(1) in the present invention, using the built-in RAID of SSD, the die where block is wiped in garbage collection procedure is regarded as of short duration
Failure, the degraded mode of similar RAID respond read request in such a way that RAID is rebuild, can effectively solve the problem that in garbage reclamation
Read conflict.
(2) in the present invention, RAID is built in interchannel, makes full use of the concurrency inside SSD, because of the data of solid-state disk
Read-write granularity is a Hash memory pages, therefore its band is organized as unit of page.
Detailed description of the invention
Fig. 1 is the multichannel SSD configuration diagram of the embodiment of the present invention；
Fig. 2 is the schematic diagram of internal structure of NAND Flash chip；
Fig. 3 is the process schematic that NAND Flash executes garbage reclamation；
Fig. 4 is the read-write requests process flow diagram of the embodiment of the present invention；
Fig. 5 is that the SSD interchannel of the embodiment of the present invention builds RAID5 operation principle schematic diagram；
Fig. 6 is in the method for the embodiment of the present invention for 4 kinds of calculating verification page mode schematic diagrames of different write requests；
The schematic diagram for rebuilding data page is brought in the method for Fig. 7 embodiment of the present invention using RAID5 same；
Fig. 8 is the method processing read request flow diagram of the embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
As shown in Figure 1, SSD is made of a series of flash arrays, there are multiple flash controllers (nfc), Mei Geshan inside SSD
Memory controller controls a channel (channel), works independently between flash controller, each channel has the channel of oneself total
Line；The multiple flash chips of carry (chip) on each channel.Fig. 2 is chip internal structure, and each flash chip is by multiple wafers
(die) it forms；Each wafer is made of multiple groupings (plane)；Each grouping has multiple pieces (block), and block is flash chip
The basic unit of erasing, each piece has multiple pages (page), and page is the basic unit of flash chip read-write.
According to the method for one embodiment of the invention, as shown in figure 5, first with data page between each channel (channel)
(page) RAID is established for band, preferably RAID5 in the present embodiment, the port number of built-in RAID5 is preferably 5, but this hair
It is not limited to this in bright, as long as the RAID that can be verified is all suitable for this.Each band is by four data pages and a verification page group
At being belonging respectively to 5 channels of composition RAID5, Ap, Bp, Cp, Dp are verification page, A0-A3, B0-B3, C0-C3, D0- in Fig. 5
D3 is data page.
For the presence of the built-in RAID of SSD, for the consistency for guaranteeing SSD data, write request per treatment is in addition to writing it
Except the data page of body, the verification data page of RAID stripe also will be updated.Calculating or updating verification data page includes entire item
With writing for the first time, data, individual data page writes data for the first time and individual data web update writes data.
Data page is write in this programme for the first time and more new data writes and whole band writes data and writes the calculating of individual data page
The mode for verifying data page is different.
As shown in fig. 6, write data for the first time for entire band, detailed process be can be such that such as Fig. 6 (A), and band is
Sky, page where data D0, D1, D2, D3 are sky at the beginning, and the page where verifying data P is also empty.Data D0, D1 is written simultaneously,
After D2, D3, verification data P is done XOR operation and is obtained by data D0, D1, D2, D3, finally by data D0, D1, D2, D3 and verification
Data P is written flash memory and constitutes a band.
As shown in fig. 6, write data for the first time for individual data page, detailed process be can be such that such as Fig. 6 (B), to certain
New data D0 is written in one band, original checksums data P is read into DRAM from flash memory first, then verifies data P and number
New verification data P ' is obtained according to D0 exclusive or, flash memory finally is written into data D0 and new verification data P '.
As shown in fig. 6, write data for individual data web update, detailed process be can be such that such as Fig. 6 (C), when making
When with new data D0 ' Lai Gengxin legacy data D0, legacy data D0 and verification data P0 are read out first from flash memory, then will
Legacy data D0 and new data D0 ' step-by-step exclusive or, the result after exclusive or is again with verification data P0 step-by-step exclusive or, obtained result P0 '
As new verification data finally write back to new data D0 ' and new verification data P0 ' in flash memory.
As shown in fig. 6, detailed process can be such that as shown in Fig. 6 (D), simultaneously for entire band more new writen data
Use new data D0 ', D1 ', D2 ', D3 ' updates legacy data D0, D1, D2, D3, the new data P ' that verifies by new data D0 ', D1 ',
D2 ', D3 ' exclusive or obtain, finally by new data D0 ', D1 ', D2 ', D3 ' and new verification data P ' write back flash memory together.
As shown in fig. 7, the process of description data reconstruction: there is read request to read data D2, D2 on channel2 at this time
Affiliated die is wiped free of operation and occupies, and short duration failure occurs, by way of the data exclusive or of other pages in RAID5 band
To restore data D2.
In the present invention, promotes SSD method of reading performance during garbage reclamation and specifically include that
In single solid-state disk SSD inside configuration, each interchannel of flash controller control is established by band of data page
RAID；
When handling read request, if the wafer (die) where the data page of read request access is not carrying out erasing behaviour
Make, at this time read request normal response；
If the wafer (die) where the data page of read request access is just being wiped free of occupied by operation, short duration failure occurs,
RAID reconstruction is carried out by reading the data of other data pages on same band using the RAID of interchannel at this time, thus and
When respond the read request.
Wherein RAID is rebuild restores the read request other data pages in RAID stripe by way of exclusive or
The data page of required reading.
Specifically, as shown in figure 8, the method for promotion SSD reading performance during garbage reclamation of one embodiment of the invention,
It is carried out by way of the built-in RAID5 response system read request of interchannel, the specific steps of which are as follows:
1. there is read request, counter i=1 is set, i, which is used to record, rebuilds the completed port number of read request；
It is occupied 2. whether the die where judging the read request is wiped free of operation, is to go to step 3., 8. no step turns；
3. judging whether i is equal to 1, it is, goes to step 4., otherwise goes to step 5.；
4. generating (n-1) a reconstruction read request to be mounted to respectively on a channel (n-1)；Read the number of remaining RAID stripe
According to n is total number of channels；
Parallel processing 5. (n-i) a channel；Successively existing read-write requests in treatment channel；
6. the respective data blocks of same band on channel are read, the reconstruction read request on one channel of every processing, i
++；
It is to turn 9., no turn 2. 7. judging whether (n-i) is equal to 0；
8. directly reading data block, the trace that RAID is rebuild is removed, is turned 10.；
9. restoring the data block to be read using RAID5 according to existing n-1 data block；
10. responding read request.
In the present solution, the request on channel is sequential processes, rebuilds read request and be also sequentially mounted to corresponding channel
On.Because existing number of request is not generally identical on each channel, the time for completing to rebuild reading is also different.Pass through wheel
The mode of inquiry is read to guarantee that all channel complete reconstruction.Because former before not completed there may be RAID reconstruction operation
The upper erasing operation of channel is completed one step ahead, does not need to carry out RAID reconstruction again at this time, needs to rebuild the raid not completed
Trace is removed and (mainly removes and rebuild read request data generated), if that is, erasing operation is complete in carrying out RAID reconstruction process
At wafer (die) can be accessed again, then eliminate the RAID and rebuild trace, normal response read request.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should all include
Within protection scope of the present invention.
Claims (8)
1. a kind of method for promoting SSD reading performance during garbage reclamation, comprising:
In single solid-state disk SSD inside configuration, RAID is established by band of data page to each interchannel of flash controller control；
When handling read request, if the wafer (die) where the data page of read request access is not carrying out erasing operation, this
When read request normal response；
If the wafer (die) where the data page of read request access is just being wiped free of occupied by operation, short duration failure occurs, at this time
RAID reconstruction is carried out by reading the data of other data pages on same band using the RAID of interchannel, to ring in time
It should read request；
Wherein, the RAID is rebuild restores the read request other data pages in RAID stripe by way of exclusive or
The data page of required reading；The RAID in each channel rebuild read operation can parallel processing, and do not influence on the channel
Normal read-write requests.
2. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 1, wherein carrying out
If the erasing operation is completed in RAID reconstruction process, i.e., wafer (die) can be accessed again, then eliminate the RAID and rebuild trace
Mark, normal response read request.
3. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 1 or 2, wherein each
The perhaps update verification data page calculating is calculated when processing write requests or update verification data page includes entire band first
Secondary data, the individual data page write writes data, individual data web update data and entire band more new data for the first time.
4. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 3, wherein described entire
Band first time data packets include while writing data to each data page, and verification page data does XOR operation by each data page data and obtains
It arrives, each data page data and verification page data is finally constituted into a band, flash memory is written.
5. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 3, wherein described single
Data page first time data packets include the write-in data of a new data page into any bar band, first by original checksums page from sudden strain of a muscle
DRAM is read in depositing, and then verifies the verification data of page and the new data page data exclusive or obtains newly verifying data, finally
Flash memory is written into the new data page data and new verification data.
6. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 3, wherein described single
Data web update data packets are included reads the former verification data of the former data of the individual data page and verification page from flash memory
Out, then by the former data and new data step-by-step exclusive or, the result after exclusive or verifies data step-by-step exclusive or with original again, obtains
Result be new verification data, finally new data and new verification data are write back in flash memory.
7. a kind of method for promoting SSD reading performance during garbage reclamation according to claim 3, wherein described entire
Band updates data packets and includes while updating each data page data using new data, verifies page data by the new of each data web update
Data exclusive or obtains, and the new data of each data web update and new verification page data are finally write back flash memory.
8. a kind of method method by the built-in RAID response system read request of interchannel, its step are as follows:
1) in single solid-state disk SSD inside configuration, each interchannel of flash controller control is established by band of data page
RAID；
2) there is read request, set counter i=1；
3) whether the wafer (die) where judging the read request, which is wiped free of operation, occupies, and is to go to step 4), 9) no step turns；
4) judge whether i is equal to 1, be, go to step 5), otherwise go to step 6)；
5) it generates n-1 reconstruction read request to be mounted to respectively on n-1 channel, reads the data of remaining RAID stripe, wherein n is
Total number of channels；
6) n-i channel of parallel processing, successively existing read-write requests in treatment channel；
7) respective data blocks of same band on channel, the reconstruction read request on one channel of every processing, i++ are read；
8) judge whether n-i is equal to 0, be, go to step 10), otherwise go to step 3)；
9) data block is directly read, the trace that RAID is rebuild is removed, goes to step 11)；
10) according to existing n-1 data block, restore the data block to be read using RAID5；
11) read request is responded.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610877009.XA CN106527971B (en) | 2016-10-08 | 2016-10-08 | A method of promoting SSD reading performance during garbage reclamation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610877009.XA CN106527971B (en) | 2016-10-08 | 2016-10-08 | A method of promoting SSD reading performance during garbage reclamation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106527971A CN106527971A (en) | 2017-03-22 |
| CN106527971B true CN106527971B (en) | 2019-11-26 |
Family
ID=58333045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610877009.XA Active CN106527971B (en) | 2016-10-08 | 2016-10-08 | A method of promoting SSD reading performance during garbage reclamation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106527971B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10453547B2 (en) * | 2017-06-16 | 2019-10-22 | Seagate Technologies Llc | Monitoring a memory for retirement |
| CN107678690A (en) * | 2017-09-29 | 2018-02-09 | 山东超越数控电子有限公司 | A kind of implementation method of solid state hard disc and its RAID array |
| CN110895445B (en) * | 2018-09-12 | 2021-09-14 | 华为技术有限公司 | Data processing method and system |
| CN110119250B (en) * | 2019-05-13 | 2023-02-10 | 湖南国科微电子股份有限公司 | Nonvolatile storage medium data processing method and nonvolatile storage medium |
| CN110780814B (en) * | 2019-10-10 | 2021-08-06 | 苏州浪潮智能科技有限公司 | Stored data sorting method and device |
| CN113495830A (en) * | 2020-04-02 | 2021-10-12 | 深圳星火半导体科技有限公司 | Method and system for testing and analyzing efficiency of garbage recovery algorithm |
| CN112527694A (en) * | 2020-12-15 | 2021-03-19 | 深圳星火半导体科技有限公司 | Fragmentation degree alarm method and device |
| CN114327287B (en) * | 2021-12-30 | 2023-08-04 | 苏州浪潮智能科技有限公司 | Method, device, equipment and medium for setting wafer of solid state disk |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336744A (en) * | 2013-06-20 | 2013-10-02 | 华中科技大学 | Garbage recovery method for solid-state storage device and system for garbage recovery method |
| CN103412826A (en) * | 2013-07-18 | 2013-11-27 | 记忆科技(深圳)有限公司 | Garbage collection method and system of solid state disk |
-
2016
- 2016-10-08 CN CN201610877009.XA patent/CN106527971B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336744A (en) * | 2013-06-20 | 2013-10-02 | 华中科技大学 | Garbage recovery method for solid-state storage device and system for garbage recovery method |
| CN103412826A (en) * | 2013-07-18 | 2013-11-27 | 记忆科技(深圳)有限公司 | Garbage collection method and system of solid state disk |
Non-Patent Citations (1)
| Title |
|---|
| GC-RAIS:一种基于垃圾回收感知的固态盘阵列;吴素贞、陈晓熹、毛波;《计算机研究与发展》;20130131;第50卷(第1期);第60-68页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106527971A (en) | 2017-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106527971B (en) | A method of promoting SSD reading performance during garbage reclamation | |
| EP2937774B1 (en) | Storage controller, storage system and method of operating storage controller | |
| CN103902465B (en) | Method and system for recycling solid state disk junk and solid state disk controller | |
| CN104503706B (en) | A kind of data storage and read method based on disk array | |
| US10942849B2 (en) | Use of a logical-to-logical translation map and a logical-to-physical translation map to access a data storage device | |
| US20150378886A1 (en) | Software-defined ssd and system using the same | |
| CN103218274B (en) | A kind of method that trouble saving is cumulative and solid state hard disc | |
| CN109358809B (en) | RAID data storage system and method | |
| CN103049222A (en) | RAID5 (redundant array of independent disk 5) write IO optimization processing method | |
| US9251059B2 (en) | Storage system employing MRAM and redundant array of solid state disk | |
| CN110347613B (en) | Method for realizing RAID in multi-tenant solid-state disk, controller and multi-tenant solid-state disk | |
| US11954332B2 (en) | Data processing method, controller, storage device, and storage system | |
| CN105912307A (en) | Data processing method and device of Flash controller | |
| US20140173223A1 (en) | Storage controller with host collaboration for initialization of a logical volume | |
| CN103823728A (en) | Intelligent reestablishing method for RAID (redundant array of independent disks) | |
| CN103534688A (en) | Data recovery method, storage equipment and storage system | |
| CN110413454B (en) | Data reconstruction method and device based on storage array and storage medium | |
| WO2021139166A1 (en) | Error page identification method based on three-dimensional flash storage structure | |
| CN103019893A (en) | Multi-disk fault-tolerant two-dimensional hybrid disk RAID4 system architecture and read-write method thereof | |
| CN102023902B (en) | Disc array reconstruction method | |
| CN105988727A (en) | RAID-based storage method and storage device | |
| US20240037119A1 (en) | A method and device for storing data in a distributed database | |
| CN108959526A (en) | Blog management method and log management apparatus | |
| CN105988939A (en) | Data storage device and flash memory control method | |
| CN101729421B (en) | Storage method and device based on time division multiplex |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |