CN103019966A - Data storage system comprising backup body memory and management method thereof - Google Patents
Data storage system comprising backup body memory and management method thereof Download PDFInfo
- Publication number
- CN103019966A CN103019966A CN2011102828639A CN201110282863A CN103019966A CN 103019966 A CN103019966 A CN 103019966A CN 2011102828639 A CN2011102828639 A CN 2011102828639A CN 201110282863 A CN201110282863 A CN 201110282863A CN 103019966 A CN103019966 A CN 103019966A
- Authority
- CN
- China
- Prior art keywords
- data
- archive memory
- storage device
- storage system
- controller
- 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
Links
- 230000015654 memory Effects 0.000 title claims abstract description 102
- 238000013500 data storage Methods 0.000 title claims abstract description 66
- 238000007726 management method Methods 0.000 title claims abstract description 39
- 230000005540 biological transmission Effects 0.000 claims description 16
- 238000004140 cleaning Methods 0.000 claims description 14
- 230000015556 catabolic process Effects 0.000 claims description 13
- 230000007717 exclusion Effects 0.000 claims description 12
- 230000008878 coupling Effects 0.000 claims 3
- 238000010168 coupling process Methods 0.000 claims 3
- 238000005859 coupling reaction Methods 0.000 claims 3
- 238000012790 confirmation Methods 0.000 abstract 1
- 230000005611 electricity Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000000034 method Methods 0.000 description 5
- 239000000835 fiber Substances 0.000 description 4
- 230000008676 import Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Abstract
The invention relates to a data storage system and a management method thereof. The data storage system comprises N storage devices, a backup memory and a controller, wherein the N is a natural number, and each storage device comprises a respective written-in cache. Once the data storage system is subjected to electricity interruption, the backup memory still can keep data stored in the backup memory. The controller is used for executing one reserved operation to multiple pieces of data conveyed from an access requirement unit so as to generate multiple pieces of written-in data. The written-in data is conveyed to the written-in caches of the N storage devices, the written-in data is copied into the backup memory, the N storage devices are responded to convey one piece of written-in confirmation message, and the written-in data copied into the backup memory is indicated as ''available''.
Description
Technical field
The present invention is about a kind of data storage system (data storage system) and management method (managing method) thereof, especially for the data storage system that meets disk array (redundant array of independent drives, RAID) framework.And especially, data storage system of the present invention comprises archive memory, and the data of wanting to write when avoiding data storage system to suffer power breakdown are vanished from sight, and protected.
Background technology
Along with user's storage data amount is increasing, the data storage system (also being called disk array (RAID) system) that meets disk array (RAID) framework is used to store mass data widely.Disk array (RAID) system can provide the data storing space of high availability (highavailability), high-effect (high performance) or large capacity (high volume) to main frame (host).
The different technologies that disc array system utilizes is called as the RAID rank, also is divided into: RAID 0, RAID 1, RAID2, RAID 3, RAID 4, RAID 5 and RAID 6.Each the each have their own advantage of RAID rank technology and shortcoming.
The formation of existing disc array system, the disk array that comprises a disk array controller (RAID controller) and consisted of by a plurality of disc drivers (disk drive).Disk array controller system is connected to each disc driver, and this disk array is defined as one or more logic magnetic disc drivers (logical disk drive) of RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 and is reached by this disk array controller.
See also Fig. 1, an existing framework that meets the data storage system 1 of a RAID framework is illustrated among Fig. 1.This data storage system 1 comprises a plurality of storage devices 12 and a controller 10.
In a specific embodiment, these a plurality of storage devices 12 are that standard becomes a disk array framework, for example, and RAID 0, RAID1, RAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 etc.
This controller 10 is to produce (reconstruction) redundant data identical with data to be read (redundant data).For example, under the RAID5 framework, this controller 10 be see through mutual exclusion or (XOR) computing produce redundant data.
In practical application, each storage device 12 can be a magnetic tape station (tape drive), a disc driver (disk drive), a memory device (memory drive), a smooth store recording device (optical storage drive), or in same disc driver corresponding to one of single read/write head sector, or other equivalent storage device.
Be shown in equally Fig. 1, this controller 10 is to be connected to each storage device 12.Fig. 1 and illustrate an access requirement unit (application I/O request unit) 2.This access requires unit 2 to be connected to this controller 10 via a transmission interface 14.In practical application, it can be network computer, mini-computer, mainframe, mobile computer that this access requires unit 2, or need any electronic equipment of the data in the reading out data stocking system 1, for example, mobile phone, personal digital assistant (personaldigital assistant, PDA), digital video recording equipment, digital music player, etc.
It is one one independently during electronic equipment when this access requires unit 2, it can see through and store LAN (SAN), LAN (LAN), sequence advanced technology (serial ATA, import and export (I/O) interfaces such as the SATA) transmission interfaces such as interface, fiber channel (FC), small computer standard interface (SCSI), or PCI Express are connected with this data storage system 1.In addition, when this access requirement unit 2 is a special integrated circuit package, or other can send the equivalent device of import and export reading requirement, it can send the data reading requirement to this controller 10 according to the order of other device (or request), and then sees through this controller 10 and read data in these storage devices 12.
This controller 10 in this data storage system 1 and these storage devices 12 can be placed in the single casing simultaneously, also can independently separate and are placed in the different casings.On the practice, this controller 10 can see through fiber channel (FC), small computer system interface (SCSI), sequence and connect the transmission interfaces such as SCSI (SAS), sequence A TA (SATA), parallel ATA (PATA, or claim IDE) and be connected with these storage devices 12.If these storage devices 12 are disc driver, each storage device 12 can be the disc driver of the various distinct interface kenels such as FC, SAS, SCSI, SATA, PATA.This controller 10 is to can be disk array controller, or has the controller of the data storage system that can produce redundant data.
Be shown in equally Fig. 1, each storage device 12 comprises separately a write cache (write cache) 122.Each storage device 12 is distinguished into a plurality of data segments (block).From fault-tolerant viewpoint, can be divided into user's data segments and coordination data segments two classes.User's data segments stores general user's data.The coordination data segments then stores one group of unnecessary coordination data, with when fault-tolerant demand is arranged, for inverse user data.Be present in corresponding user's data segments of different storing devices and coordination data segments and form one and store segmentation band (stripe), wherein the system of the data in the coordination data segments carries out mutual exclusion or (XOR) result of computing gained by the data in user's data segments.What must state is, those of ordinary skill in the art all understands, the calculating of the data in the coordination data segments, except using mutual exclusion or (XOR) the computing, also comprise coordination miscellaneous (parity) computing or similar computing, as long as there is following relation: the data in any data segments in the same stored segmentation band can be calculated and be got by the data of its corresponding data segments.
The user holds wish that data are write 1 li of this data storage system, and this controller 10 receives many data that this access requires unit 2 to transmit via this transmission interface 14.Then, this controller 10 is carried out a predetermined operation (for example, mutual exclusion or (XOR) computing) to produce many data writings with these data that receive.Then, this controller 10 transmits these data writings to the write cache 122 of these a plurality of storage devices 12.These data writings be stored to the write cache 122 of these a plurality of storage devices 12 interior after, these a plurality of storage devices 12 are replied these controllers 10 1 and are write the affirmation message.At this moment, this controller 10 assert that namely these data writings have write in these a plurality of storage devices 12, and can not protect these data.But, be stored in the recording medium (media) that these data writings in the write cache 122 of these a plurality of storage devices 12 may not write this a plurality of storage device 12 itself totally.In case during the power breakdown that this data storage system 1 suffers not expect, these data writings that still are stored in the write cache 122 of these a plurality of storage devices 12 namely can be vanished from sight.
Summary of the invention
Therefore, the invention provides a kind of data storage system and management method thereof.And especially, the present invention's data storage system comprises archive memory, and the data of wanting to write when avoiding data storage system to suffer power breakdown are vanished from sight, and protected.
The data storage system of preferred embodiment one of according to the present invention, it comprises N storage device, archive memory and a controller, and wherein N is a natural number.Each storage device comprises separately a write cache.In case when this data storage system suffered power breakdown, this archive memory was still preserved storage data within it.Many the data that this controller transmits in order to receive an access requirement unit, these data that receive are carried out a predetermined operation to produce many data writings, transmit these data writings to the write cache of this N storage device, copy these data writings to this archive memory, and respond this N storage device and transmit one and write and confirm that these data writings that message will be copied in this archive memory are denoted as effectively.And, judging the storage area of this archive memory when this controller is lower than a threshold value or copies the address of these data writings to this archive memory when repeating, this controller issue one fast command fetch of cleaning (flush cache command) is to this N storage device, and responds this N storage device and transmit one and clear up cache and finish the message removing and be stored in the interior data writing of this archive memory.
In a specific embodiment, when this data storage system start, this controller judges whether there is the effective data writing of sign in this archive memory.If judged result is for affirming, this controller writes this N storage device with indicating effective data writing in this archive memory.And, these being indicated after effective data writing writes this N storage device finishing, this controller is removed the data writing that is stored in this archive memory.
In a specific embodiment, this predetermined operation can be a mutual exclusion or (XOR) computing, a mirror image computing (mirroringoperation), mathematics maximum distance separable codes (maximum distance separable code) computing without computing (NO operation), all minutes computing (slice operation), the essential multinomial operation in a limited Patrick Gallois field (Galois field, GF), a finite field, (parity) computing of other coordination or similar computing.
In a specific embodiment, this archive memory is the volatile memory of a non-voltile memory or a configuration one independent current source.
In a specific embodiment, this archive memory system only is electrically connected to this controller.In another specific embodiment, this archive memory is write cache one of in this controller.In another specific embodiment, this archive memory system is connected electrically between this controller and this N the storage device.
The management method of preferred embodiment is for the management method of a data storage system one of according to the present invention.This data storage system comprises a plurality of storage devices and an archive memory.Each storage device comprises separately a write cache.When this data storage system suffered power breakdown, this archive memory was still preserved storage data within it.The management method of the preferred embodiment according to the present invention at first, receives many data that an access requirement unit transmits.Then, the management method according to the present invention is carried out a predetermined operation to produce many data writings with these data that receive.Then, the management method according to the present invention judges whether the storage area of this archive memory is lower than a threshold value, or judges whether copy these data writings to the address in this archive memory repeats.If the determination result is NO the person of deciding, the manager's rule according to the present invention transmits these data writings to the write cache of this N storage device, and copies these data writings to this archive memory.And management method according to the present invention responds this N storage device transmission one and writes the affirmation message, and these data writings that are copied in this archive memory are denoted as effectively.
If the storage area of this archive memory is lower than a threshold value or copies these data writings to the address in this archive memory and repeat, the fast command fetch of manager's rule issue one cleaning according to the present invention is to this N storage device.And management method according to the present invention responds this N storage device transmission one cleaning cache and finishes message, removes the data writing that is stored in this archive memory.
When this data storage system start, the management method according to the present invention judges whether there is the effective data writing of sign in this archive memory.If judged result is the yea-sayer, manager's rule according to the present invention writes this N storage device with indicating effective data writing in this archive memory, and these are indicated after effective data writing writes this N storage device finishing, remove the data writing that is stored in this archive memory.
Advantage about the present invention can be by following embodiment and appended graphic being further understood with spirit.
Description of drawings
The invention will be further described below in conjunction with accompanying drawing:
Fig. 1 is an existing configuration diagram that meets the data storage system of RAID framework.
Fig. 2 is the configuration diagram of the one preferred embodiment data storage system according to the present invention.
Fig. 3 is the synoptic diagram of according to a preferred embodiment of the present invention one of its archive memory of data storage system distortion.
Fig. 4 is the synoptic diagram of another distortion of its archive memory of data storage system according to a preferred embodiment of the present invention.
Fig. 5 is the process flow diagram of the one preferred embodiment management method according to the present invention.
Fig. 6 is another process flow diagram of the preferred embodiment management method according to the present invention.
The component symbol explanation:
[primary clustering symbol description]
1,3: data storage system 10,30: controller
12,32: storage device 122,322: write cache
14,34: transmission interface 36: archive memory
2,4: access requires unit 5: management method
S50～S70: process step
Embodiment
The invention provides a kind of data storage system and management method thereof.And especially, data storage system comprises archive memory according to the present invention, and the data of wanting to write when avoiding data storage system to suffer power breakdown are vanished from sight, and protected.Below by the detailed description to preferred embodiment of the present invention, use abundant explanation about feature of the present invention, spirit, advantage and the feasibility of implementing.
See also Fig. 2, the framework of the data storage system 3 of preferred embodiment is illustrated among Fig. 2 one of according to the present invention.This data storage system 3 comprises N storage device 32, a controller 30 and an archive memory 36, wherein N be a natural number (1,2,3 ... ..).
In a specific embodiment, this N storage device 32 standards become a disk array framework, for example, and RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 etc.
This controller 30 is to produce (reconstruction) redundant data identical with data to be read.For example, under RAID 5 frameworks, this controller 30 be see through mutual exclusion or (XOR) computing produce redundant data.
In a specific embodiment, each storage device 32 can be a magnetic tape station, a disc driver, a memory device, a smooth store recording device, or in same disc driver corresponding to one of single read/write head sector, or other equivalent storage device.
Be shown in equally Fig. 2, this controller 30 is connected to each storage device 32.Fig. 2 and illustrate an access requirement unit 4.This access requires unit 4 to be connected to this controller 30 via a transmission interface 34.In practical application, it can be network computer, mini-computer, mainframe, mobile computer that this access requires unit 4, or need any electronic equipment of the data in the reading out data stocking system 3, for example, mobile phone, personal digital assistant, digital video recording equipment, digital music player, etc.
It is one one independently during electronic equipment when this access requires unit 4, it is can see through to store LAN (SAN), LAN (LAN), sequence advanced technology (serial ATA, import and export (I/O) interfaces such as the SATA) transmission interfaces such as interface, fiber channel (FC), small computer standard interface (SCSI), or PCI Express are connected with this data storage system 3.In addition, when this access requirement unit 4 is a special integrated circuit package, or other can send the equivalent device of import and export reading requirement, it can send the data reading requirement to this controller 30 according to the order of other device (or request), and then sees through this controller 30 and read data in these storage devices 32.
This controller 30 in this data storage system 3 and these storage devices 32 can be placed in the single casing simultaneously, also can independently separate and are placed in the different casings.On the practice, this controller 30 can see through fiber channel (FC), small computer system interface (SCSI), sequence and connect the transmission interfaces such as SCSI (SAS), sequence A TA (SATA), parallel ATA (PATA, or claim IDE) and be connected with these storage devices 32.If these storage devices 32 are disc driver, each storage device 32 can be the disc driver of the various distinct interface kenels such as FC, SAS, SCSI, SATA, PATA.This controller 30 is to can be disk array controller, or has the controller of the data storage system that can produce redundant data.
Be shown in equally Fig. 2, each storage device 32 comprises separately a write cache 322.Each storage device 32 is distinguished into a plurality of data segments.From fault-tolerant viewpoint, can be divided into user's data segments and coordination data segments two classes.User's data segments stores general user's data.The coordination data segments then stores one group of unnecessary coordination data, with when fault-tolerant demand is arranged, for inverse user data.Be present in corresponding user's data segments of different storing devices and coordination data segments and form one and store segmentation band, wherein the system of the data in the coordination data segments carries out mutual exclusion or (XOR) result of computing gained by the data in user's data segments.What must state is, those of ordinary skill in the art all understands, the calculating of the data in the coordination data segments, except using mutual exclusion or (XOR) the computing, for various RAID frameworks, it also can be the mirror image computing, without computing (NO operation), the cutting computing, limited Patrick Gallois field (Galois field, GF) essential multinomial operation, the mathematics maximum distance separable codes of finite field (maximum distance separable code) computing, other is with bit arithmetic or similar computing, as long as there is following relation: the data in any data segments in the same stored segmentation band can be calculated and be got by the data of its corresponding data segments.
The user holds wish that data are write 3 li of this data storage systems, and this controller 30 receives many data that this access requires unit 4 to transmit via this transmission interface 34.Then, this controller 30 is carried out a predetermined operation (for example, mutual exclusion or (XOR) computing) to produce many data writings with these data that receive.Then, this controller 30 transmits these data writings to the write cache 322 of this N storage device 32.Especially, this controller 30 copies these data writings to this archive memory 36.
These data writings be stored to the write cache 322 of this N storage device 32 interior after, this N storage device 32 is replied these controllers 30 1 and is write the affirmation message.At this moment, writing of these controller 30 this N of response storage device 32 transmission confirms that these data writings that message will be copied in this archive memory 36 are denoted as effectively, still protect these data writings.But, be stored in these data writings in the write cache 322 of this N storage device 32 and may not have write in the recording medium of this N storage device 32 own totally.
And, judge the storage area of these archive memories 36 when this controller 30 and be lower than a threshold value or copy these data writings to the address in this archive memory 36 when repeating that these controller 30 issues one are cleared up fast command fetch to this N storage device 32.After this N storage device 32 receives that this clears up fast command fetch, namely the data writings in these write caches 322 can be write in the recording medium of this N storage device 32 own.After these data writings in these high-speed caches 322 write the recording medium of this N storage device 32 itself totally, this N storage device 32 was namely replied this controller 30 1 and is cleared up caches and finish message.At this moment, this controller 30 namely responds this cleaning cache and finishes message, removes the data writing that is stored in this archive memory 36.
Especially, in case during the power breakdown that this data storage system 3 suffers not expect, this archive memory 36 is still preserved the data that store within it.When this data storage system 3 start, this controller 30 judges whether there is the effective data writing of sign in this archive memory 36.If judged result is for affirming that this controller 30 writes this N storage device 32 with the effective data writing of these archive memory 36 interior signs.And, these being indicated after effective data writing writes this N storage device 32 finishing, this controller 30 is removed the data writing that is stored in this archive memory 36.By this, the data of wanting to write when the data storage system 3 according to the present invention can be avoided suffering power breakdown are vanished from sight, and guarantee that the data of wanting to write write in this N storage device 32.
In a specific embodiment, this archive memory 36 is a non-voltile memory (such as Flash Memory), or the volatile memory of configuration one independent current source (for example, battery or UPS).
In a specific embodiment, as shown in Figure 2, this archive memory 36 only is electrically connected to this controller 30.
In another specific embodiment, as shown in Figure 3, this archive memory 36 is write cache one of in this controller 30.Have the assembly with Fig. 2 duplicate numbers mark among Fig. 3, identical or similar function and running are arranged, do not do at this and give unnecessary details.
In another specific embodiment, as shown in Figure 4, this archive memory 36 is for being connected electrically between this controller 30 and this N the storage device 32.Have the assembly with Fig. 2 duplicate numbers mark among Fig. 4, identical or similar function and running are arranged, do not do at this and give unnecessary details.
See also Fig. 5, Fig. 5 is the process flow diagram that one of illustrates according to the present invention the management method 2 of preferred embodiment.Management method 5 according to the present invention is the management method for for example Fig. 2, Fig. 3 or data storage system 3 shown in Figure 4.The framework of this data storage system 3 in above describing in detail, does not repeat them here.
As shown in Figure 5, according to management method 5 of the present invention execution in step S50 at first, receive many data that an access requirement unit 4 transmits.
Then, according to management method 5 execution in step S52 of the present invention, these data that receive are carried out a predetermined operation, to produce many data writings.
In a specific embodiment, this predetermined operation can be a mutual exclusion or (XOR) computing, a mirror image computing, a mathematics maximum distance separable codes computing without computing, all minutes computing, the essential multinomial operation in a limited Patrick Gallois field (Galois field, GF), a finite field, other is with bit arithmetic or similar computing.
Then, according to management method 5 execution in step S54 of the present invention, judge whether the storage area of this archive memory 36 is lower than a threshold value.
If step S54 is the determination result is NO the person of deciding, according to management method 5 execution in step S56 of the present invention, judge whether copy these data writings to addresses in this archive memory 36 repeats.
If the determination result is NO the person of deciding of step S56 according to management method 5 execution in step S58 of the present invention, transmits these data writings to the write cache 322 of this N storage device 32, and copies these data writings to this archive memory 36.
After step S58, according to management method 5 execution in step S60 of the present invention, respond 32 transmission one of this N storage device and write the affirmation message, these data writings that are copied in this archive memory 36 are denoted as effectively.
If the judged result of step S54 is the yea-sayer, or the judged result of step S56 is the yea-sayer, the management method 5 execution in step S62 according to the present invention, and the fast command fetch of issue one cleaning is to this N storage device 32.At this moment, after these data writings in these high-speed caches 322 write the recording medium of this N storage device 32 itself totally, this N storage device 32 was namely replied one and is cleared up cache and finish message.
After step S62, according to management method 5 execution in step S64 of the present invention, this cleaning cache that responds 32 transmission of this N storage device is finished message, removes the data writing that is stored in this archive memory 36.After step S64, namely finish program.
See also Fig. 6, when this data storage system 3 start, according to management method 5 further execution in step S66 of the present invention, judge whether there is the effective data writing of sign in this archive memory 36.
If the judged result of step S66 is the yea-sayer, according to management method 5 further execution in step S68 of the present invention, the effective data writing of these archive memory 36 interior signs is write this N storage device 32.After step S68, namely finish these are indicated effective data writing write after this N storage device, according to management method 5 execution in step S70 of the present invention, removing the data writing that is stored in this archive memory 36.By this, the data of wanting to write when management method 5 according to the present invention can avoid this data storage system 3 to suffer power breakdown are vanished from sight, and guarantee that the data of wanting to write write in this N storage device 32.
By the detailed description of above the present invention's preferred embodiment, can know and understand data storage system provided by the present invention and management method thereof that the data of wanting to write in the time of avoiding data storage system to suffer power breakdown are vanished from sight, and protected.
By the detailed description of above preferred embodiment, hope can be known the feature and spirit of describing the present invention more, and is not to come the present invention's category is limited with above-mentioned disclosed preferred embodiment.On the contrary, its objective is that hope can contain in the category of claim of being arranged in of various changes and tool equality institute of the present invention wish application.Therefore, the category of the claim that the present invention applies for should be done the broadest explanation according to above-mentioned explanation, contains the arrangement of all possible change and tool equality to cause it.
Claims (12)
1. a data storage system requires the unit to become electrically coupling with an access, comprises:
N storage device optionally exported one and write the affirmation message, and each storage device comprises separately a write cache, and N is a natural number;
One archive memory is in case when this data storage system suffered power breakdown, this archive memory was still preserved storage data within it; And
One controller, in order to carrying out a predetermined operation to produce many data writings from many data that this access requires the unit to transmit, transmit these data writings to the write cache of this N storage device, copy these data writings to this archive memory, and respond this and write the affirmation message, these data writings that are copied in this archive memory are denoted as " effectively ".
2. data storage system according to claim 1, it is characterized in that: this N storage device is optionally exported a cleaning cache and is finished message, when the storage area of this archive memory is lower than a threshold value or copies the address of these data writings to this archive memory when repeating, the fast command fetch of this controller issue one cleaning is to this N storage device, and respond this cleaning cache and finish message, remove the data writing that is stored in this archive memory.
3. data storage system according to claim 2, it is characterized in that: when this data storage system start, if there is sign in this archive memory " effectively " data writing, this controller then should write back this N storage device by effective data writing, and after finishing, remove the data writing that is stored in this archive memory.
4. data storage system according to claim 3 is characterized in that: this predetermined operation is selected from the one in the group that is comprised of without the mathematics maximum distance separable codes computing of computing, all minutes computing, limited Patrick Gallois field essence multinomial operation and a finite field a mutual exclusion or (XOR) computing, a mirror image computing.
5. data storage system according to claim 3, it is characterized in that: this archive memory is the volatile memory of a non-voltile memory or a configuration one independent current source, this archive memory only is electrically connected to the write cache in this controller, this controller, or is connected electrically between this controller and this N the storage device.
6. management method for a data storage system, this data storage system requires the unit to become electrically coupling with an access, and comprise N storage device and an archive memory, N is a natural number, each storage device comprises separately a write cache, in case when this data storage system suffered power breakdown, this archive memory was still preserved storage data within it, this management method comprises the following step:
(a) receive many data that this access requirement unit transmits;
(b) data that these received are carried out a predetermined operation to produce many data writings;
Whether the storage area of (c) judging this archive memory is lower than a threshold value, or judges whether copy these data writings to the address in this archive memory repeats;
If step (c) the determination result is NO the person of deciding carries out the following step:
Transmit these data writings to the write cache of this N storage device, and copy these data writings to this archive memory; And
Respond one of this N storage device transmission and write the affirmation message, these data writings that will be copied in this archive memory are denoted as " effectively ".
7. management method according to claim 6, if step (c) judged result is the yea-sayer, carry out the following step:
The fast command fetch of issue one cleaning is to this N storage device; And
Respond one of this N storage device transmission cleaning cache and finish message, remove the data writing that is stored in this archive memory.
8. management method according to claim 7 is characterized in that: when this data storage system start, carry out the following step:
Judge and whether have sign in this archive memory " effectively " the data that writes; And
If judged result is the yea-sayer, writes this N storage device with indicating effective data writing in this archive memory, and after finishing, remove the data writing that is stored in this archive memory.
9. described management method according to claim 8 is characterized in that: this predetermined operation is selected from the one in the group that is comprised of without the mathematics maximum distance separable codes computing of computing, all minutes computing, limited Patrick Gallois field essence multinomial operation and a finite field a mutual exclusion or (XOR) computing, a mirror image computing.
10. described management method according to claim 9, it is characterized in that: this archive memory is the volatile memory of a non-voltile memory or a configuration one independent current source.
11. a data storage system requires the unit to become electrically coupling with an access, comprises:
N storage device optionally exported one and write the affirmation message, and each storage device comprises separately a write cache, and N is a natural number;
One archive memory is in case when this data storage system suffered power breakdown, this archive memory was still preserved storage data within it; And
One controller, in order to carrying out a predetermined operation to produce many data writings from many data that this access requires the unit to transmit, copy these data writings to this archive memory, and respond this and write the affirmation message, these data writings that are copied in this archive memory are denoted as " effectively ".
12. data storage system according to claim 11, it is characterized in that: this controller transmits these data writings to the write cache of this N storage device, wherein, this N storage device and optionally export one the cleaning cache finish message, when the storage area of this archive memory is lower than a threshold value or copies the address of these data writings to this archive memory when repeating, the fast command fetch of this controller issue one cleaning is to this N storage device, and respond this cleaning cache and finish message, removing is stored in the data writing in this archive memory, wherein when this data storage system start, if there is sign in this archive memory " effectively " data writing, this controller then should write back this N storage device by effective data writing, and after finishing, remove the data writing that is stored in this archive memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110282863.9A CN103019966B (en) | 2011-09-22 | 2011-09-22 | A kind of data storage system and management method thereof comprising backup body internal memory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110282863.9A CN103019966B (en) | 2011-09-22 | 2011-09-22 | A kind of data storage system and management method thereof comprising backup body internal memory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103019966A true CN103019966A (en) | 2013-04-03 |
| CN103019966B CN103019966B (en) | 2015-08-19 |
Family
ID=47968591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110282863.9A Expired - Fee Related CN103019966B (en) | 2011-09-22 | 2011-09-22 | A kind of data storage system and management method thereof comprising backup body internal memory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103019966B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6243795B1 (en) * | 1998-08-04 | 2001-06-05 | The Board Of Governors For Higher Education, State Of Rhode Island And Providence Plantations | Redundant, asymmetrically parallel disk cache for a data storage system |
| US20060174156A1 (en) * | 2005-02-01 | 2006-08-03 | Sridhar Balasubramanian | Cache redundancy for lsi raid controllers |
| CN101521039A (en) * | 2008-02-29 | 2009-09-02 | 群联电子股份有限公司 | Data storage system, controller and method therefor |
| TW200947205A (en) * | 2008-05-13 | 2009-11-16 | Jmicron Technology Corp | Raid5 controller and accessing method with data stream distribution and aggregation operations based on the primitive data access block of storage devices |
-
2011
- 2011-09-22 CN CN201110282863.9A patent/CN103019966B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6243795B1 (en) * | 1998-08-04 | 2001-06-05 | The Board Of Governors For Higher Education, State Of Rhode Island And Providence Plantations | Redundant, asymmetrically parallel disk cache for a data storage system |
| US20060174156A1 (en) * | 2005-02-01 | 2006-08-03 | Sridhar Balasubramanian | Cache redundancy for lsi raid controllers |
| CN101521039A (en) * | 2008-02-29 | 2009-09-02 | 群联电子股份有限公司 | Data storage system, controller and method therefor |
| TW200947205A (en) * | 2008-05-13 | 2009-11-16 | Jmicron Technology Corp | Raid5 controller and accessing method with data stream distribution and aggregation operations based on the primitive data access block of storage devices |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103019966B (en) | 2015-08-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102945201B (en) | The non-volatile media log recording of verification msg set | |
| CN102981922B (en) | Select be used for data repository repeat delete agreement | |
| CN102681794B (en) | Method and system for realizing redundant array protection of a disk based on double controllers | |
| CN103718162B (en) | Method and apparatus for flexible RAID in SSD | |
| CN101916173B (en) | RAID (Redundant Array of Independent Disks) based data reading and writing method and system thereof | |
| CN103534688B (en) | Data reconstruction method, memory device and storage system | |
| CN103513942B (en) | The reconstructing method of raid-array and device | |
| JPWO2006123416A1 (en) | Disk failure recovery method and disk array device | |
| CN101984400B (en) | RAID control method, device and system | |
| CN102799533B (en) | Method and apparatus for shielding damaged sector of disk | |
| EP2573689A1 (en) | Method and device for implementing redundant array of independent disk protection in file system | |
| CN101980137A (en) | Method, device and system for reconstructing redundant array of inexpensive disks | |
| CN103336727B (en) | The method that NAND flash memory storage equipment is carried out data manipulation | |
| CN103699457A (en) | Method and device for restoring disk arrays based on stripping | |
| CN101556802B (en) | RAID array transposing method and device thereof | |
| CN105302665A (en) | Improved copy-on-write snapshot method and system | |
| CN102135862B (en) | Disk storage system and data access method thereof | |
| CN107885620B (en) | Method and system for improving performance and reliability of solid-state disk array | |
| CN107728943B (en) | Method for delaying generation of check optical disc and corresponding data recovery method | |
| US20180107546A1 (en) | Data storage system with virtual blocks and raid and management method thereof | |
| CN102193848B (en) | Data recovery method and device for bad sector of logic unit | |
| CN102495680A (en) | Reconstruction method of RAID (Redundant Array of Independent Disks) system | |
| CN104572374A (en) | Storage processing method, storage processing device and storage equipment | |
| US8898392B2 (en) | Data storage system including backup memory and managing method thereof | |
| CN106933707B (en) | Data recovery method and system of data storage device based on raid technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20150819 Termination date: 20170922 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |