CN103713973B - A kind of mixing storage backup method and system based on HDD and SSD - Google Patents
A kind of mixing storage backup method and system based on HDD and SSD Download PDFInfo
- Publication number
- CN103713973B CN103713973B CN201410009126.5A CN201410009126A CN103713973B CN 103713973 B CN103713973 B CN 103713973B CN 201410009126 A CN201410009126 A CN 201410009126A CN 103713973 B CN103713973 B CN 103713973B
- Authority
- CN
- China
- Prior art keywords
- data
- hdd
- segment
- ssd
- write
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000013507 mapping Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 238000000151 deposition Methods 0.000 claims 1
- 230000007246 mechanism Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 101100045541 Homo sapiens TBCD gene Proteins 0.000 description 1
- 101150093640 SSD1 gene Proteins 0.000 description 1
- 101100111629 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR2 gene Proteins 0.000 description 1
- 102100030290 Tubulin-specific chaperone D Human genes 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Techniques For Improving Reliability Of Storages (AREA)
Abstract
The present invention provides a kind of mixing storage backup method and system based on HDD and SSD, and applied to Computer Applied Technology field, the above method comprises the following steps:Single or multiple HDD linear address space is divided into some equal portions;Wherein, each equal portions include data segment and free segment;If write request reaches, the HDD is write data into corresponding data segment according to preset strategy.By implementing technical scheme, the data on SSD can be backed up in time with HDD, the mixed structure realizes a kind of low price, high performance reliable memory system data back-up.
Description
Technical field
The present invention relates to Computer Applied Technology field, more particularly to a kind of mixing storage backup side based on HDD and SSD
Method and system.
Background technology
In recent years, the solid-state disk based on flash memory (SSD) because its superior performance widely paid close attention to and using work as
In preceding storage system backup, still, due to restricted lifetime, SSD reliability enjoys query;HDD array (RAID) is a kind of
Traditional is used for improving the means of reliability, but is not particularly suited for SSD.First, RAID mechanism needs extra memory space to protect
Check information is deposited, but system cost can be dramatically increased by being equipped with extra SSD storage check informations;Secondly, traditional RAID machines
System is tended to balancedly be distributed write request between each disk, and in the RAID based on SSD, this way causes each SSD life-span
Equably be lost, when a certain SSD due to the life-span exhausts and when breaking down, other SSD also reach lifetime limitation, so as to
There is polydisc failure, trigger loss of data.
At this stage, traditional RAID mechanism main storage system backup and SSD are primarily present following 3 aspects in terms of read-write
The problem of:
1st, in the RAID based on SSD, because the same type of SSD life-spans are substantially consistent, each SSD life-span is equably
Loss, when a certain SSD due to the life-span exhausts and when breaking down, other SSD also reach lifetime limitation, more so as to occur
Disk failure, trigger loss of data.
2nd, RAID mechanism needs extra memory space to preserve check information, and compared to HDD, SSD prices are still expensive, match somebody with somebody
Standby extra SSD storage check informations can dramatically increase system cost.
3rd, traditional HDD can not meet the needs of user in read or write speed, and the mixing of the characteristic of present SSD high speeds makes
With being constantly present some shortcomings in performance.
The content of the invention
The present invention provides a kind of mixing storage backup method and system based on HDD and SSD, to solve the above problems.
The present invention provides a kind of mixing storage backup method based on HDD and SSD.The above method includes:
Single or multiple HDD linear address space is divided into some equal portions, each of which equal portions include data segment and
Free segment, the data segment are used for the data for backing up SSD, and the free segment is used for temporarily preserving data;
If write request reaches, the HDD is write data into corresponding data segment according to preset strategy.
The present invention also provides a kind of mixing storage and backup system based on HDD and SSD, including:One or more HDD, one
Individual or multiple SSD, nonvolatile memory and host side;
The SSD is used to respond read request and write request that the host side is sent;
HDD and nonvolatile memory the composition backup module, the write request sent for responding host side, wherein:
The nonvolatile memory be used for bridging HDD and SSD delay difference away from;
The linear address space of the HDD is divided into some equal portions;Each of which equal portions include data segment and free segment,
The data segment and free segment are alternately arranged on linear address space, and the data segment is used for the data for backing up the SSD, institute
State SSD address to correspond with the address of data segment on the HDD, the free segment is used for temporarily preserving data;
The HDD is used for when write request reaches, and is write data into according to preset strategy in corresponding data segment.
By implementing technical scheme, the data on SSD can be backed up in time with HDD, the mixed structure is real
A kind of low price, high performance reliable memory system data back-up are showed;The present invention is using nonvolatile memory bridge joint HDD and SSD
Between gaps between their growth rates, substantially increase response speed and performance requirement of the storage system backup to the read-write requests of user;This
Invention additionally uses a kind of I/O restructuring strategy, and the strategy can strengthen the succession of HDD built-in functions, make HDD suitable as far as possible
Sequence is read and write, and strengthens HDD write performance;In addition, the present invention adopts to reduce the situation of the burst write request occurred on monolithic SSD
With polylith HDD backup polylith SSD, ensure that the burst on a certain SSD writes data and can disperse to backup in multiple HDD in time, it is real
A kind of low price, high performance reliable memory system backup are showed.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, forms the part of the application, this hair
Bright schematic description and description is used to explain the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 show the general frame schematic diagram of the mixing storage system back mechanism of embodiments of the invention 1;
Fig. 2 show the I/O restructuring schematic diagrames in the HDD of embodiments of the invention 2;
Fig. 3 show the polylith HDD collaboration backup configuration diagrams of embodiments of the invention 3;
Fig. 4 show the storage backup method process chart of the mixing based on HDD and SSD of embodiments of the invention 4.
Embodiment
Describe the present invention in detail below with reference to accompanying drawing and in conjunction with the embodiments.It should be noted that do not conflicting
In the case of, the feature in embodiment and embodiment in the application can be mutually combined.
A kind of mixing storage backup method based on HDD and SSD is present embodiments provided, including:
Single or multiple HDD linear address space is divided into some equal portions, each of which equal portions include data segment and
Free segment, the data segment are used for the data for backing up SSD, and the free segment is used for temporarily preserving data;
If write request reaches, the HDD is write data into corresponding data segment according to preset strategy.
Wherein, the data segment and free segment are alternately arranged on linear address space, the SSD addresses and the HDD
The address of upper data segment corresponds.
Wherein, if write request reaches, the HDD according to preset strategy write data into corresponding to process in data segment
For:
When write request reaches, if data can not be write in corresponding data segment, the sky nearest from magnetic head is write data into
In not busy section, to avoid magnetic head from dispatching, while record the data using a mapping table being stored in Nonvolatile memory and write
The purpose free segment entered;
Data are read in nonvolatile memory from the free segment during HDD free time;
Data in nonvolatile memory are eventually written in corresponding data segment by the HDD.
Wherein, if the data can not be write in corresponding data segment, the free segment nearest from magnetic head is write data into
On, including:If it is different data segments that client, which sends multiple write requests and target, the multiple write request is merged,
Write in the same free segment nearest from magnetic head.
Wherein, it is described to read data in nonvolatile memory from the free segment, including:By in multiple continuous free segments
Data simultaneously read in the nonvolatile memory.
Wherein, for a specific thread, it sends multiple write requests within a period of time, and these data are written to together
In one free segment, or it is written in adjacent several free segments;
It is described to read data in nonvolatile memory from the free segment, including:By same free segment or several continuous
Data in free segment read internal memory;
The data by nonvolatile memory are eventually written in corresponding data segment, including:Found out in internal memory described
Multiple write requests of particular thread, and these requests are reverted into a long sequence.
Wherein, using polylith HDD backup polylith SSD, HDD quantity is selected according to HDD and SSD amount of capacity;
The space of oneself is divided into some equal portions by any HDD according to SSD quantity, is respectively supplied to every piece of SSD;SSD
The data of oneself are divided into some equal portions according to HDD quantity, backuped to respectively on every piece of HDD.
Fig. 1 show the general frame schematic diagram of the mixing storage system back mechanism of embodiments of the invention 1, such as Fig. 1
It is shown, one piece of SSD is backed up with one piece of HDD, the read request that host side is sent only is responded by SSD, and write request is by SSD and backup module
Respond simultaneously;Contain a piece of nonvolatile memory and HDD in backup module, nonvolatile memory is used for bridging HDD and SSD delay difference
Away from;Inside HDD, it is proposed that a kind of I/O recombinates tactful (description of detailed in Example 2) to strengthen the succession of operation, makes number
According to HDD can be backuped in time.
Fig. 2 show the I/O restructuring schematic diagrames in the HDD of embodiments of the invention 2, the I/O scheduling mechanisms in monolithic HDD
As shown in Figure 2:In Fig. 2 (a), HDD linear address space is divided into many sections, wherein, it is data segment labeled as D,
It is that free segment, data segment and free segment are linearly spatially being alternately arranged labeled as F, there is a free time behind each two data segment
Section (just having a free segment in real system, behind dozens of data segment), SSD data backup is on data segment, free segment
Then it is used for temporarily preserving data.The mechanism is designed with two features:1) data segment is regularly distributed over HDD addresses with free segment
Spatially, data segment is used for backing up SSD data, and the address space of data segment composition and SSD address space correspond, and
This corresponding relation can be calculated directly, without extra address mapping table;2) in the mechanism, near HDD optional positions
A free time all be present, when write request reaches, no matter magnetic head is in the where of linear address space, a sky nearby all be present
Not busy section, data can be write in the free segment in time, and any write request can not have to scheduling magnetic head and sequentially complete.
Fig. 2 (b) describes strategy caused by this order write request, and a write request should write D1, but magnetic head is present
D2 positions are in, in order to avoid magnetic head is dispatched, data are not written directly in D1, on the contrary, we find the sky nearest from magnetic head
Not busy section F3, is write data into wherein, and so, data segment D1 data are written into free segment F3, it is necessary to safeguard a token record
This corresponding relation;Free segment is interim preservation user data, and when the HDD free time, user data is transferred to data segment
On, I/O regrouping process can be summarized as 3 steps in HDD exemplified above:1. write request is write data into from magnetic when reaching
In nearest free segment;2. data are read in nonvolatile memory from free segment during the HDD free time;3. by nonvolatile memory
Data be eventually written in corresponding data segment, the I/O operation that this 3 steps are related to all is order.In the present embodiment on
I/O restructuring strategies in HDD are mainly realized ordered by two kinds of means:It is the scheduling for reducing magnetic head first;Next to that will be more
Individual short request is merged into a long request, realizes that step is:(1) if data can be write in corresponding data segment, due to SSD
The address one-to-one corresponding of the upper data segment of address and HDD, it is not necessary to which mapping table records this corresponding relation;If on the contrary, data
It can not write in corresponding data segment, then need to write in the free segment nearest from magnetic head, at this time, it may be necessary to which a mapping table is remembered
Which these data are recorded to write in free segment (mapping table is stored in nonvolatile memory):When write request reaches, data are write
Onto the free segment nearest from magnetic head, a magnetic head scheduling is avoided, client often sends multiple write requests in a short time,
If these write requests are packaged into a long request, write in same free segment, can effectively reduce the number of write operation,
Strengthen the succession of write operation.As shown in Fig. 2 (c), client sends two write requests, request 1 and request 2, their target
For different data segments, now magnetic head is in D2 positions, and this strategy merges the two write requests, writes free segment F3 (from magnetic
Nearest free segment) in, the capacity of each section is larger (being set as 2MB in the present invention), so, dozens of user's write request
One long request can be packaged into write in same free segment;(2) data in free segment are read into internal memory:Multiple user's requests
After merging is write in same free segment, these data can all be read by a read operation, reduce read operation
Number;In addition, adjacent several free segments also can together be read, the probability of magnetic head scheduling, example so can be further reduced
Such as:3 adjacent free segment F1, F2, F3 are read simultaneously, because these free segments are sequentially distributed in HDD spatially, magnetic head
Scheduling can avoid;(3) data in internal memory are write back into HDD:Data in multiple continuous free segments are read internal memory simultaneously, this
A little data are eventually write in respective data segment, and these data sequentially can be write in HDD;For a specific thread,
It may send multiple write requests within a period of time, and these data are written in same free segment, or be written to adjacent
In several free segments.
In above-mentioned steps, the data in several continuous free segments are read into internal memory, it is specific that some can be found out in internal memory
Multiple write requests of thread, and these requests are reverted into a long sequence, this process is had been illustrated in Fig. 2 (d), and 3 are write
Request, request 1, request 2 and request 3, wherein request 1 and request 3 belong to same continuous sequence, but requested 2 interrupt, 3
Request is all written in free segment F3, and when the data in F3 are read internal memory, the succession of request 1 and request 3 is restored,
So when the data in internal memory are write back into HDD, multiple write requests can be merged into a long sequence, reduce write operation.
Coordination mechanism between a kind of polylith HDD of design of the embodiment of the present invention is selected according to HDD and SSD amount of capacity
HDD quantity is selected, the space of oneself is divided into some equal portions according to SSD quantity, is respectively supplied to every piece of SSD by any HDD;
The data of oneself are divided into some equal portions by SSD according to HDD quantity, are backuped to respectively on every piece of HDD, and the mechanism can will be specific
Burst write request on SSD is distributed on polylith HDD, so as to effectively suppress the load peak on single HDD.
The space that optimisation strategy has monolithic HDD of writing inside monolithic HDD is much larger than monolithic SSD, causes the wave in HDD spaces
Take;And performance gap between HDD and SSD be present, if upper layer host sends substantial amounts of burst write request, HDD can not be timely
These data are handled, cause mass data accumulation in internal memory, if memory pressure is excessive, SSD must stop responding user's
Write request, in order to solve this problem, the embodiment of the present invention devises the coordination mechanism between a kind of polylith HDD, such as Fig. 3 institutes
Show, 2 pieces of HDD provide backup for 3 pieces of SSD, and every piece of HDD is divided into 3 parts, back up the data on 3 pieces of SSD, each portion respectively
Divide and address space is divided into data segment and free segment, data segment provides backup for specific SSD, in Fig. 3, labeled as D1's
Data segment only uses for SSD1, is only used labeled as D2 data segment for SSD2, and free segment is shared for all SSD, so, nothing
By magnetic head be in where, can all be backuped in time in the free segment near magnetic head for any SSD write request, this hair
Multiple HDD of bright embodiment cooperation can handle the burst write request for some SSD calmly.
Fig. 4 show the storage backup method process chart of the mixing based on HDD and SSD of embodiments of the invention 4, bag
Include following steps:
Step 401:Single or multiple HDD linear address space is divided into some equal portions;Each of which equal portions include
Data segment and free segment, the data segment are used for the data for backing up SSD, and the free segment is used for temporarily preserving data;
Step 402:If write request reaches, the HDD is write data into corresponding data segment according to preset strategy.
The present embodiment additionally provides a kind of mixing storage and backup system based on HDD and SSD, it is characterised in that including:One
Individual or multiple HDD, one or more SSD, nonvolatile memory and host side;
The SSD is used to respond read request and write request that the host side is sent;
HDD and nonvolatile memory the composition backup module, the write request sent for responding host side, wherein:
The nonvolatile memory be used for bridging HDD and SSD delay difference away from;
The linear address space of the HDD is divided into some equal portions;Each of which equal portions include data segment and free segment,
The data segment and free segment are alternately arranged on linear address space, and the data segment is used for the data for backing up the SSD, institute
State SSD address to correspond with the address of data segment on the HDD, the free segment is used for temporarily preserving data;
The HDD is used for when write request reaches, and is write data into according to preset strategy in corresponding data segment.
Wherein, the space of oneself is divided into some equal portions by any HDD according to SSD quantity in the HDD, is provided respectively
To every piece of SSD;
The HDD is write data into corresponding data segment according to preset strategy, including:If data can not be write accordingly
In data segment, then write data into the free segment nearest from magnetic head, to avoid magnetic head from dispatching, while use be stored in it is non-volatile
Property internal memory in a mapping table record the purpose free segment of data write-in;
The HDD is additionally operable at one's leisure read data in nonvolatile memory from the free segment, then will be non-volatile
Data in internal memory are eventually written in corresponding data segment.
Wherein, the HDD is write data into corresponding data segment according to preset strategy, including:Sent in client more
Individual write request and when target is different data segment, the multiple write request is merged, writes the same sky nearest from magnetic head
In not busy section;
The HDD reads data in nonvolatile memory from the free segment, including:By in multiple continuous free segments
Data are read in the nonvolatile memory simultaneously.
By implementing technical scheme, the data on SSD can be backed up in time with HDD, the mixed structure is real
A kind of low price, high performance reliable memory system data back-up are showed;The present invention is using nonvolatile memory bridge joint HDD and SSD
Between gaps between their growth rates, substantially increase response speed and performance requirement of the storage system backup to the read-write requests of user;This
Invention additionally uses a kind of I/O restructuring strategy, and the strategy can strengthen the succession of HDD built-in functions, make HDD suitable as far as possible
Sequence is read and write, and strengthens HDD write performance;In addition, the present invention adopts to reduce the situation of the burst write request occurred on monolithic SSD
With polylith HDD backup polylith SSD, ensure that the burst on a certain SSD writes data and can disperse to backup in multiple HDD in time, it is real
A kind of low price, high performance reliable memory system backup are showed.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.
Claims (9)
- A kind of 1. mixing storage backup method based on HDD and SSD, it is characterised in that including:Single or multiple HDD linear address space is divided into some equal portions, and each of which equal portions include data segment and free time Section, the data segment are used for the data for backing up SSD, and the free segment is used for temporarily preserving data;If write request reaches, the HDD is write data into corresponding data segment according to preset strategy;The space of oneself is divided into some equal portions by any HDD according to SSD quantity in the HDD, is respectively supplied to every piece SSD;The data of oneself are divided into some equal portions by SSD according to HDD quantity, are backuped to respectively on every piece of HDD;If write request reaches, the process that the HDD is write data into corresponding data segment according to preset strategy is:If data can not be write in corresponding data segment, write data into the free segment nearest from magnetic head, specifically include:If It is different data segments that client, which sends multiple write requests and target, then merges the multiple write request, write from magnetic head In nearest same free segment.
- 2. according to the method for claim 1, it is characterised in thatThe data segment and free segment are alternately arranged on linear address space, the SSD addresses and data segment on the HDD Address corresponds.
- 3. according to the method for claim 2, it is characterised in that if write request reaches, the HDD will according to preset strategy Process corresponding to data write-in in data segment is:When writing data into the free segment nearest from magnetic head, to avoid magnetic head from dispatching, while use is stored in non-volatile A mapping table in depositing records the purpose free segment of the data write-in;Data are read in nonvolatile memory from the free segment during HDD free time;Data in nonvolatile memory are eventually written in corresponding data segment by the HDD.
- 4. according to the method for claim 3, it is characterised in thatIt is described to read data in nonvolatile memory from the free segment, including:Data in multiple continuous free segments are same When read in the nonvolatile memory.
- 5. according to the method for claim 3, it is characterised in thatFor a specific thread, it sends multiple write requests within a period of time, and these data are written to same free segment It is interior, or be written in adjacent several free segments;It is described to read data in nonvolatile memory from the free segment, including:By same free segment or several continuous idle Data in section read internal memory;The data by nonvolatile memory are eventually written in corresponding data segment, including:Found out in internal memory described specific Multiple write requests of thread, and these requests are reverted into a long sequence.
- 6. according to the method described in claim 1 or 2 or 3, it is characterised in thatUsing polylith HDD backup polylith SSD, HDD quantity is selected according to HDD and SSD amount of capacity.
- A kind of 7. mixing storage and backup system based on HDD and SSD, it is characterised in that including:One or more HDD, one or Multiple SSD, nonvolatile memory and host side;The SSD is used to respond read request and write request that the host side is sent;HDD and nonvolatile memory the composition backup module, the write request sent for responding host side, wherein:The nonvolatile memory be used for bridging HDD and SSD delay difference away from;The linear address space of the HDD is divided into some equal portions;Each of which equal portions include data segment and free segment, described Data segment and free segment are alternately arranged on linear address space, and the data segment is used for the data for backing up the SSD, described SSD address and the address of data segment on the HDD correspond, and the free segment is used for temporarily preserving data;The HDD is used for when write request reaches, and is write data into according to preset strategy in corresponding data segment;The space of oneself is divided into some equal portions by any HDD according to SSD quantity in the HDD, is respectively supplied to every piece SSD;The data of oneself are divided into some equal portions by SSD according to HDD quantity, are backuped to respectively on every piece of HDD;The HDD is write data into corresponding data segment according to preset strategy, including:If data can not write corresponding data Duan Zhong, then write data into the free segment nearest from magnetic head, specifically include:Multiple write requests and target are sent in client For different data segments when, the multiple write request is merged, write in the same free segment nearest from magnetic head.
- 8. mixing storage and backup system according to claim 7, it is characterised in thatThe HDD is write data into corresponding data segment according to preset strategy, in addition to:It is nearest from magnetic head when writing data into Free segment on, to avoid magnetic head from dispatching, while the number is recorded using a mapping table being stored in Nonvolatile memory According to the purpose free segment of write-in;The HDD is additionally operable at one's leisure read data in nonvolatile memory from the free segment, then by nonvolatile memory In data be eventually written in corresponding data segment.
- 9. mixing storage and backup system according to claim 8, it is characterised in thatThe HDD reads data in nonvolatile memory from the free segment, including:By the data in multiple continuous free segments Read simultaneously in the nonvolatile memory.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410009126.5A CN103713973B (en) | 2014-01-08 | 2014-01-08 | A kind of mixing storage backup method and system based on HDD and SSD |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410009126.5A CN103713973B (en) | 2014-01-08 | 2014-01-08 | A kind of mixing storage backup method and system based on HDD and SSD |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103713973A CN103713973A (en) | 2014-04-09 |
CN103713973B true CN103713973B (en) | 2017-11-17 |
Family
ID=50406974
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410009126.5A Active CN103713973B (en) | 2014-01-08 | 2014-01-08 | A kind of mixing storage backup method and system based on HDD and SSD |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103713973B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10346245B2 (en) | 2014-12-09 | 2019-07-09 | Tsinghua University | Data storage system and data storage method |
CN104778018B (en) * | 2015-04-23 | 2018-06-05 | 南京道熵信息技术有限公司 | Wide band disk array and storage method based on asymmetric hybrid magnetic disk mirroring |
CN106775453B (en) * | 2016-11-22 | 2019-07-05 | 华中科技大学 | A kind of construction method mixing storage array |
CN106708433A (en) * | 2016-12-06 | 2017-05-24 | 深圳市深信服电子科技有限公司 | Read-write method and device for stored data |
CN108874594B (en) * | 2018-06-25 | 2021-10-22 | 郑州云海信息技术有限公司 | Disk replacement method, device, equipment and storage medium in storage cluster |
CN109032513B (en) * | 2018-07-16 | 2020-08-25 | 山东大学 | RAID (redundant array of independent disks) architecture based on SSD (solid State disk) and HDD (hard disk drive) and backup and reconstruction methods thereof |
CN109597579A (en) * | 2018-12-03 | 2019-04-09 | 郑州云海信息技术有限公司 | The method that tactful configuration is carried out to extended chip on board and rear end disk |
CN110286858A (en) * | 2019-06-26 | 2019-09-27 | 北京奇艺世纪科技有限公司 | A kind of data processing method and relevant device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102591593A (en) * | 2011-12-28 | 2012-07-18 | 华为技术有限公司 | Method for switching hybrid storage modes, device and system |
CN103150128A (en) * | 2013-03-25 | 2013-06-12 | 中国人民解放军国防科学技术大学 | Implementation method of solid state drive (SSD) and disk-based reliable mixed storage system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102136274B (en) * | 2009-12-30 | 2013-03-27 | 爱国者电子科技有限公司 | Mobile hard disk with two storage media |
US8909859B2 (en) * | 2012-03-01 | 2014-12-09 | HGST Netherlands B.V. | Implementing large block random write hot spare SSD for SMR RAID |
CN202795333U (en) * | 2012-07-09 | 2013-03-13 | 董锴 | Magnetic disk redundancy array high-speed read-write control circuit structure in server |
CN103076993A (en) * | 2012-12-28 | 2013-05-01 | 北京思特奇信息技术股份有限公司 | Storage system and method for concentration type system |
-
2014
- 2014-01-08 CN CN201410009126.5A patent/CN103713973B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102591593A (en) * | 2011-12-28 | 2012-07-18 | 华为技术有限公司 | Method for switching hybrid storage modes, device and system |
CN103150128A (en) * | 2013-03-25 | 2013-06-12 | 中国人民解放军国防科学技术大学 | Implementation method of solid state drive (SSD) and disk-based reliable mixed storage system |
Also Published As
Publication number | Publication date |
---|---|
CN103713973A (en) | 2014-04-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103713973B (en) | A kind of mixing storage backup method and system based on HDD and SSD | |
US7971013B2 (en) | Compensating for write speed differences between mirroring storage devices by striping | |
US7634618B2 (en) | Methods, systems, and computer program products for optimized copying of logical units (LUNs) in a redundant array of inexpensive disks (RAID) environment using buffers that are smaller than LUN delta map chunks | |
US9229653B2 (en) | Write spike performance enhancement in hybrid storage systems | |
US8909887B1 (en) | Selective defragmentation based on IO hot spots | |
CN101436151B (en) | Data real time backup method and system based on file system | |
US7321955B2 (en) | Control device, control method and storage medium recording a control program for controlling write-back schedule of data from cache memory to a plurality of storage devices | |
WO2018019119A1 (en) | Method and device for dynamic partial-parallel data layout for continuous data storage | |
US8650379B2 (en) | Data processing method for nonvolatile memory system | |
CN100524237C (en) | Computer hard disk data continuous time point storage and recovery method | |
CN102043687B (en) | Copy on first write device for realizing data snapshot and control method | |
US20060010290A1 (en) | Logical disk management method and apparatus | |
CN103049220B (en) | Storage controlling method, memory control device and solid-state memory system | |
CN103136074A (en) | Data storage method and data storage system of multiple disk array systems | |
CN107357680A (en) | Data processing equipment and the method for data processing | |
CN101815989A (en) | Capturing and naming dynamic storage tiering configurations to support data pre-staging | |
CN102201005A (en) | Method for distributing streaming media enhanced file of file system | |
CN104699416A (en) | Data storage system and data storage method | |
CN103823728A (en) | Intelligent reestablishing method for RAID (redundant array of independent disks) | |
CN106970765A (en) | Date storage method and device | |
US11093339B2 (en) | Storage utilizing a distributed cache chain and a checkpoint drive in response to a data drive corruption | |
US8922930B1 (en) | Limit disc nodes by band usage | |
US20090083503A1 (en) | System of creating logical volume and method thereof | |
JP2006079274A (en) | File management device, network system, file management method and program | |
CN102750111B (en) | A kind of method of RAID bitmap storage |
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 |