CN113535082A - Method for realizing wear inverse equilibrium - Google Patents
Method for realizing wear inverse equilibrium Download PDFInfo
- Publication number
- CN113535082A CN113535082A CN202110640205.6A CN202110640205A CN113535082A CN 113535082 A CN113535082 A CN 113535082A CN 202110640205 A CN202110640205 A CN 202110640205A CN 113535082 A CN113535082 A CN 113535082A
- Authority
- CN
- China
- Prior art keywords
- wear
- raid5
- reverse
- data
- disk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000002441 reversible effect Effects 0.000 claims abstract description 74
- 230000005012 migration Effects 0.000 claims abstract description 20
- 238000013508 migration Methods 0.000 claims abstract description 20
- 239000007787 solid Substances 0.000 claims abstract description 19
- 230000008569 process Effects 0.000 claims description 26
- 230000007246 mechanism Effects 0.000 claims description 20
- 238000005299 abrasion Methods 0.000 claims description 7
- 230000015556 catabolic process Effects 0.000 claims description 5
- 238000006731 degradation reaction Methods 0.000 claims description 5
- 230000000593 degrading effect Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
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
- G06F3/0614—Improving the reliability of storage systems
- G06F3/0616—Improving the reliability of storage systems in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]
-
- 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
- G06F3/0629—Configuration or reconfiguration of storage systems
-
- 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
Abstract
The invention discloses a method for realizing wear inverse balance, which comprises the following steps: s1, creating RAID 5; s2, judging whether all member disks of RAID5 are solid state disks; s3, if yes, acquiring the service life of the member disk; s4, delaying, and calculating the consumption rate of the hard disk life for the consumption degree of the hard disk life before and after delaying; s5, judging whether the RAID5 is degraded or not, namely judging whether the service life of a disk in the RAID5 is exhausted or not; if yes, S9RAID5 reconstruction is carried out; s6, if not, calculating the life consumption rate of the member disk; s7, judging the wear reverse balance starting condition, if the difference value between the service life of a member disk and the service life of other hard disks is smaller than a set risk threshold value and the hard disk is close to the old age, namely higher than a preset service life threshold value, considering that the safety of RAID5 is reduced, and starting the wear reverse balance to increase the service life difference value among the hard disks is needed to ensure that only no more than one hard disk in RAID5 is in the old age; s8, wear reverse balancing, including data migration in wear reverse balancing; s9, RAID5 reconstruction.
Description
Technical Field
The invention belongs to the technical field of hard disk management, and relates to a method for realizing wear reverse balance.
Background
The conventional method for implementing wear reverse balancing is to select a hard disk or a data unit with a longer remaining life from a storage resource pool to reconstruct RAID5 when a plurality of hard disks in the hard disks that constitute RAID5 or the hard disks to which the data block unit belongs are in an old age period in which the life is about to be consumed (at this time, it may happen that more than one hard disk in RAID5 is consumed in a short time to cause data loss). The method needs to reconstruct the whole RAID5, consumes a great deal of time in the process, generates a great deal of extra reading and writing and consumes the service life of the hard disk; in addition, a proper amount of non-old disks are required to be introduced in the reconstruction process, and when most of hard disks in the storage resource pool are in the old age, the method cannot maintain RAID5 wear reverse balance.
The prior art has at least the following disadvantages:
1) it is often necessary to wait for the multiple hard disks in RAID5 to run the reverse wear mechanism until they are older.
2) When the wear reverse balance mechanism takes effect, the whole RAID5 needs to be reconstructed to ensure that the condition that the service lives of a plurality of hard disks are exhausted in a short time does not occur in RAID5, a large amount of time is consumed in the reconstruction process, and the service lives of the hard disks are lost due to a large amount of meaningless hard disk reading and writing.
3) When reconstructing RAID5, hard disks in non-old age must be used as substitutes, and the conditions are severe; and the residual life of the old hard disk under replacement is not easy to reasonably use, thus causing waste.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method for implementing wear reverse equalization, comprising the following steps:
s1, creating RAID 5;
s2, judging whether all member disks of RAID5 are solid state disks, if not, ending the abrasion reverse balancing step;
s3, if yes, acquiring the service life of the member disk;
s4, delaying, and calculating the consumption rate of the hard disk life for the consumption degree of the hard disk life before and after delaying;
s5, judging whether the RAID5 is degraded or not, namely judging whether the service life of a disk in the RAID5 is exhausted or not; if yes, S9RAID5 reconstruction is carried out;
s6, if not, calculating the member disk life consumption rate, wherein the member disk life consumption rate is obtained by calculating the hard disk life and the interval time obtained twice, and is used as the adjustment basis of the next time delay duration and as the input of the wear reverse balance, and accordingly, the wear reverse balance strategy is made;
s7, judging the wear reverse balance starting condition, knowing the current remaining life condition and life consumption rate of each member disk, if the difference value between the life of the member disk and the life of other hard disks is smaller than a set risk threshold value and the hard disk is close to the old age, namely higher than a preset life threshold value, considering that the security of RAID5 is reduced, and needing to start the wear reverse balance to increase the life difference value between the hard disks, so as to ensure that only no more than one hard disk in RAID5 is in the old age; if not, the abrasion reverse balance is not required to be started, and the step returns to S3;
s8, wear reverse balancing, including data migration in wear reverse balancing;
s9, reconstructing RAID5, when the service life of a member disk in RAID5 is exhausted, RAID5 enters a degradation state and reminds a system administrator to replace a hard disk, and at the moment, if the data of the member disk needing to be read is disconnected, other data blocks in the same strip are read and target data are obtained through calculation; and after the member disk is replaced, the RAID5 reconstruction process is carried out, the data is obtained through calculation and stored in a new member disk, and after the reconstruction is finished, the data migration generated in the process of the wear reverse balance mechanism is recovered.
Preferably, the creating RAID5 includes a read-write counting function for the check block P, as a basis for selecting a migration target data block during data migration.
Preferably, the obtaining of the service life of the member disk is specifically to record the whole disk writing times of the solid state disk in a main control chip of the solid state disk, and the obtaining is checked through a smartclt tool under a linux system.
Preferably, the delay is set to an initial value, and the delay time is dynamically adjusted in subsequent steps according to the life consumption rate.
Preferably, the determination of whether RAID5 is degrading does not trigger wear leveling and prompts a system administrator to replace the hard disk with an exhausted life.
Preferably, the wear reverse balancing is to select the data block D in the hard disk with the highest life span and exchange data with the check block P in the hard disk with the next highest life span in the same stripe.
Preferably, the wear reverse balancing includes copying the data of the data block D to the storage space to which the check block P belongs, modifying the stripe bit table, that is, exchanging the addresses originally pointing to the data block D and the check block P, and finally calculating to generate the information of the check block P and storing the information in the corresponding storage address.
Preferably, the wear reverse balancing comprises copying the entire data stripe to other free stripes that do not store data, and changing the data distribution during the copying.
Preferably, the steps of S8 and S9 are both followed by returning to the step of S3.
Preferably, the member disk of RAID5 corresponding to the method is divided into a plurality of data blocks, one data block is taken from each member disk to form a data stripe, one stripe is composed of N-1 data blocks D and 1 check block P, the data block D is used for storing actual data, and the check block P is used for storing check code data obtained after xor operation of the N-1 data blocks.
The beneficial effects of the invention at least comprise:
1) the operation mechanism adopts a strategy for preventing RAID5 from entering an unsafe state, and measures are taken when RAID5 enters the unsafe state to restore RAID5 to be safe, so that the safety of RAID5 is further improved.
2) Extra non-old-period hard disks are not introduced in the operation process of the wear reverse-balancing mechanism, and the operation condition is loose; the hard disk service life is planned in advance without adopting a strategy of replacing the old-age hard disk, so that the service life resources of each hard disk in the old age are reasonably utilized, and the waste of the service life of the hard disk is avoided.
3) The method utilizes the read-write characteristics of partial data to adjust the service life consumption rate of the hard disk in the RAID5, does not need to carry out integral reconstruction on the RAID5, keeps the RAID5 available online in the whole process, and generates less extra read-write quantity in the process.
Drawings
Fig. 1 is a structure diagram of a corresponding RAID5 hard disk in the method of implementing wear reverse balancing according to the embodiment of the present invention;
FIG. 2 is a state transition diagram of a method for implementing wear reverse equalization according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating steps of a method for implementing wear reverse equalization according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of data migration of a method for implementing wear-leveling according to an embodiment of the present invention;
fig. 5 is a schematic diagram of data block exchange of a method for implementing wear-leveling according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
On the contrary, the invention is intended to cover alternatives, modifications, equivalents and alternatives which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of the present invention, certain specific details are set forth in order to provide a better understanding of the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without these specific details.
Referring to fig. 1, a diagram of a wear reverse equalization RAID5 architecture is shown. The basic structure of the method is based on the implementation principle of RAID5, and because the wear reverse balancing function is the risk problem caused by the limited service life of the solid state disk in RAID5, all the member disks forming RAID5 in the implementation process of the method must adopt the solid state disks with the same capacity, otherwise, the wear reverse balancing mechanism loses the original meaning.
The member disks forming the RAID5 are divided into a plurality of data blocks, one data block is taken from each member disk to form a data strip, and one strip is composed of N-1 data blocks D (used for storing actual data) and 1 check block P (used for storing check code data obtained after XOR operation of the N-1 data blocks). In order to maximize the utilization of the external interface bandwidth of each member disk, logically continuous storage spaces such as D1, D2 and D3 … … are uniformly distributed on each member disk, and the check blocks P1, P2 and P3 … … also adopt the distribution mode. Because most continuous data are uniformly stored in different member disks, only a small amount of smaller data cannot be uniformly distributed (the occupied space is smaller than the size of one strip), so that each member disk achieves certain load balancing and wear balancing effects.
In order to prevent the situation that the wear balance causes the service life of more than one hard disk in the same RAID5 to be exhausted, a certain wear reverse balance mechanism needs to be introduced to improve the safety of the whole RAID5 and avoid the situation that data cannot be recovered through verification, and the wear reverse balance mechanism is realized in a traditional RAID management module.
Referring to fig. 2, a state transition diagram is shown.
One, conventional RAID5, no wear reverse leveling mechanism, RAID5 in this state has the same data distribution as conventional RAID 5.
1. When the wear reverse-balancing program detects that the RAID5 is composed of solid state disks, the wear reverse-balancing program is operated, and a wear reverse-balancing monitoring state is entered.
2. When the service life of a certain hard disk is exhausted, the RAID5 enters a degraded state, which is the same as the traditional RAID5 mechanism.
And secondly, monitoring the wear inverse balance, wherein the wear inverse balance program is in a cycle of acquiring the service life of the hard disk, delaying and acquiring the service life of the hard disk in the state, and the service life consumption rate of the hard disk can be obtained through calculation.
3. The wear reverse-balancing program comprehensively judges that the RAID5 achieves the wear reverse-balancing effective condition by combining the current service life of the hard disk and the service life consumption rate of the hard disk, the wear reverse-balancing program performs migration operation on data in the RAID5, and then the state enters a wear reverse-balancing effective state.
And 4, the service life of one hard disk in the RAID5 is exhausted, and the state of RAID5 degradation is entered.
And thirdly, wear reverse balance takes effect, and at the moment, the wear reverse balance program changes the RAID5 data distribution through data migration, so that the purpose of adjusting the service life consumption rate of the hard disk is achieved.
5. And (4) adjusting by a wear reverse-balancing mechanism, judging that the RAID5 does not reach a wear reverse-balancing condition by the program, recovering the initial data distribution state by the RAID5, and re-entering the monitoring state by the wear reverse-balancing program.
6. At this time, the service life of the hard disk in RAID5 is exhausted, and a RAID5 degradation state is entered.
And fourthly, degrading the RAID5, wherein when the service life of one hard disk in the RAID5 is exhausted and is in a failure state, the failed hard disk needs to be replaced. If RAID5 was previously in a state of wear-reversal equilibrium effect, i.e., RAID5 data distribution was changed, then the data distribution needs to be restored after hard disk replacement.
7. The replacement hard disk is a solid state hard disk, and the abrasion reverse balancing program enters a monitoring state.
8. The hard disk is replaced by a non-solid state disk, the wear reverse balance mechanism loses the original meaning at the moment, and the RAID5 is in a traditional RAID5 state without wear balance.
Referring to fig. 3, a flowchart of a method for implementing wear reverse equalization according to an embodiment of the present invention includes the following steps:
s1, creating RAID 5;
s2, judging whether all member disks of RAID5 are solid state disks, if not, ending the abrasion reverse balancing step;
s3, if yes, acquiring the service life of the member disk;
s4, delaying, and calculating the consumption rate of the hard disk life for the consumption degree of the hard disk life before and after delaying;
s5, judging whether the RAID5 is degraded or not, namely judging whether the service life of a disk in the RAID5 is exhausted or not; if yes, S9RAID5 reconstruction is carried out;
s6, if not, calculating the member disk life consumption rate, wherein the member disk life consumption rate is obtained by calculating the hard disk life and the interval time obtained twice, and is used as the adjustment basis of the next time delay duration and as the input of the wear reverse balance, and accordingly, the wear reverse balance strategy is made;
s7, judging the wear reverse balance starting condition, knowing the current remaining life condition and life consumption rate of each member disk, if the difference value between the life of the member disk and the life of other hard disks is smaller than a set risk threshold value and the hard disk is close to the old age, namely higher than a preset life threshold value, considering that the security of RAID5 is reduced, and needing to start the wear reverse balance to increase the life difference value between the hard disks, so as to ensure that only no more than one hard disk in RAID5 is in the old age; if not, the abrasion reverse balance is not required to be started, and the step returns to S3;
s8, wear reverse balancing, including data migration in wear reverse balancing;
s9, reconstructing RAID5, when the service life of a member disk in RAID5 is exhausted, RAID5 enters a degradation state and reminds a system administrator to replace a hard disk, and at the moment, if the data of the member disk needing to be read is disconnected, other data blocks in the same strip are read and target data are obtained through calculation; and after the member disk is replaced, the RAID5 reconstruction process is carried out, the data is calculated and stored in a new member disk, and after the reconstruction is finished, the data migration generated in the process of the wear reverse balance mechanism is recovered, namely the data migration is recovered to the data arrangement state of the traditional RAID5, and if the data arrangement state is not influenced by the wear reverse balance mechanism, the process is omitted.
S1 creates RAID5, which includes read-write counting function for check block P as the basis for selecting the migration target data block during data migration.
S3, obtaining the service life of the member disk, specifically recording the whole disk writing times of the solid state disk in the main control chip of the solid state disk, and checking the whole disk writing times through a smartcll tool under the linux system.
S4, setting an initial value for the time delay, and dynamically adjusting the time delay according to the life consumption rate in the subsequent steps.
S5 judges whether RAID5 is degrading, does not trigger wear reverse balance, reminds system administrator to replace hard disk with exhausted service life.
S8 wear reverse balance, specifically, selecting the data block D in the hard disk with the highest service life, and exchanging data with the check block P in the hard disk with the next highest service life in the same stripe.
S8 wear reverse balancing includes copying data of data block D to storage space of check block P, modifying the stripe bit table, exchanging the addresses of data block D and check block P, calculating to generate check block P information and storing in corresponding storage address.
S8 wear-reverse balancing includes copying the entire data stripe to other free stripes that have no data stored, changing the data distribution during the copying process.
After the steps of S8 and S9, the process returns to the step of S3.
The data migration process is shown in fig. 4, that is, the data block D in the hard disk with the highest life span is selected, and data is exchanged with the check block P in the same stripe and belonging to the hard disk with the second highest life span. The main process of data exchange of the stripe 1 is as shown in fig. 5, copying the data of the data block D to the storage space to which the check block P belongs, modifying the stripe bit table (for storing the block address information of each data block), i.e. exchanging the addresses originally pointing to the data block D and the check block P, and finally calculating to generate the information of the check block P and storing the information in the corresponding storage address.
The explanation for this exchange process to achieve the wear-leveling effect is as follows:
1. in the same stripe, the parity block P needs to be recalculated and written when any data block D is updated, and in stripe 1 of fig. 4, the number of writes of the parity block P is the sum of the number of writes of the data blocks D1, D2, and D3. Therefore, the number of write requests of the parity block P is much higher than that of the other data blocks D in the stripe.
2. Due to the use characteristics of a storage medium flash memory in the solid state disk, the service life of the solid state disk is prolonged along with the increase of data writing times, and the service life of the solid state disk is not influenced by a reading request, so that the service life consumption rate of the hard disk can be increased by increasing the proportion of the writing request in a certain hard disk request.
3. In the same stripe, when reading of a certain data block D in a stripe fails, target data can be calculated by reading information of other data blocks D and the check block P. The read request is generated for the check block P when the RAID5 is in a degraded state or when some hard disk data is lost, and the abnormal state is only generated for a few time periods in the whole use process of the RAID 5. Therefore, under normal use conditions, the number of read requests of the parity block P is much smaller than that of the other data blocks D.
Based on the principle, the write request proportion of the hard disk with the highest service life is increased after the data block exchange operation, and the service life consumption rate of the hard disk is improved; the write request proportion of the hard disk with the next highest service life is reduced, and the service life consumption rate of the hard disk is reduced. The method achieves the aim of increasing the service life difference between the hard disks and realizes the wear inverse balance. Meanwhile, the exchange operation reduces the read request proportion of the hard disk with the highest service life, increases the read request proportion of the hard disk with the second service life, enables the total read-write requests of the two hard disks to be kept close, and maintains the load balancing effect among the hard disks in the RAID 5.
In a specific embodiment, the S8 wear leveling program may also modify the hard RAID card firmware code to implement the hard RAID card with wear leveling.
Part of hard disk manufacturers support reading the life percentage of the hard disk directly from the solid state disk, and the data can also be used as a condition for judging the effectiveness of the wear reverse balance mechanism.
The data exchange operation in the wear reverse equalization effective process can also copy the whole data stripe to other idle stripes which do not store data, and the data distribution is changed in the copying process.
The method divides the service life of the hard disk according to time periods and dynamically plans, and the specific action of the wear reverse balancing mechanism is to migrate data blocks with different read-write characteristics in RAID5, so that the process can adjust the service life consumption rate of the hard disk in RAID5 and maintain the load balance among the hard disks of RAID 5.
In the wear reverse balancing process, the RAID5 is in a readable and writable state all the time, and the online use state of the RAID5 is not influenced.
The method divides the service life of the hard disk into a plurality of segments, monitors and counts the service life consumption rate of the hard disk before the hard disk enters the old age, takes measures to change the service life consumption rate of the hard disk, realizes inverse wear balance by taking a prevention as a main strategy, and ensures that the RAID5 does not enter the unsafe state that a plurality of hard disks are in the old age at the same time.
In the operation process of the wear reverse balancing mechanism, only a part of data blocks in RAID5 are subjected to copy migration operation, more data blocks receiving write requests are migrated to a certain hard disk, so that the service life consumption rate of the hard disk can be accelerated by a small amount without influencing the read-write performance of the whole RAID5, and the use performance of RAID5 is not influenced by the small amount of data migration. And this process may keep RAID5 in online use without having to rebuild RAID5 offline.
The operation mechanism of the method can continuously monitor the current service life and the service life consumption rate of the hard disks in the RAID5, plan the time period of each hard disk in the RAID5 entering the old age, and adjust the service life consumption rate of the hard disks through a dynamic monitoring-transferring mechanism, so that the hard disks can finally reach the old age in the estimated time and only one hard disk in the RAID5 is ensured to be in the old age.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method of achieving wear reverse equalization, comprising the steps of:
s1, creating RAID 5;
s2, judging whether all member disks of RAID5 are solid state disks, if not, ending the abrasion reverse balancing step;
s3, if yes, acquiring the service life of the member disk;
s4, delaying, and calculating the consumption rate of the hard disk life for the consumption degree of the hard disk life before and after delaying;
s5, judging whether the RAID5 is degraded or not, namely judging whether the service life of a disk in the RAID5 is exhausted or not; if yes, S9RAID5 reconstruction is carried out;
s6, if not, calculating the member disk life consumption rate, wherein the member disk life consumption rate is obtained by calculating the hard disk life and the interval time obtained twice, and is used as the adjustment basis of the next time delay duration and as the input of the wear reverse balance, and accordingly, the wear reverse balance strategy is made;
s7, judging the wear reverse balance starting condition, knowing the current remaining life condition and life consumption rate of each member disk, if the difference value between the life of the member disk and the life of other hard disks is smaller than a set risk threshold value and the hard disk is close to the old age, namely higher than a preset life threshold value, considering that the security of RAID5 is reduced, and needing to start the wear reverse balance to increase the life difference value between the hard disks, so as to ensure that only no more than one hard disk in RAID5 is in the old age; if not, the abrasion reverse balance is not required to be started, and the step returns to S3;
s8, wear reverse balancing, including data migration in wear reverse balancing;
s9, reconstructing RAID5, when the service life of a member disk in RAID5 is exhausted, RAID5 enters a degradation state and reminds a system administrator to replace a hard disk, and at the moment, if the data of the member disk needing to be read is disconnected, other data blocks in the same strip are read and target data are obtained through calculation; and after the member disk is replaced, the RAID5 reconstruction process is carried out, the data is obtained through calculation and stored in a new member disk, and after the reconstruction is finished, the data migration generated in the process of the wear reverse balance mechanism is recovered.
2. The method for implementing wear reverse balancing according to claim 1, wherein the creating RAID5 includes a read-write counting function for the parity block P as a basis for selecting a migration target data block during data migration.
3. The method for realizing wear reverse balance according to claim 1, wherein the obtaining of the life of the member disk is specifically to record the number of times of full disk writing of the solid state disk in a main control chip of the solid state disk, and the life is checked by a smartcll tool under a linux system.
4. The method of claim 1, wherein the delay is set to an initial value, and the delay time is dynamically adjusted in subsequent steps according to the rate of life consumption.
5. The method for implementing wear reverse leveling of claim 1, wherein determining whether RAID5 is degrading does not trigger wear reverse leveling and prompts a system administrator to replace an out-of-life hard disk.
6. Method for implementing wear-leveling according to claim 1, wherein the wear-leveling is performed by selecting a data block D in the hard disk with the highest lifetime and exchanging data with a check block P in the same hard disk with the next highest lifetime.
7. The method for implementing wear reverse balancing according to claim 1, wherein the wear reverse balancing includes copying data of the data block D to a storage space to which the parity block P belongs, modifying the local stripe bit table, that is, exchanging addresses originally pointing to the data block D and the parity block P, and finally calculating to generate parity block P information and storing the parity block P information at a corresponding storage address.
8. The method for implementing wear reverse leveling of claim 1, wherein the wear reverse leveling includes copying an entire data stripe to other free stripes that do not store data, and changing data distribution during the copying.
9. The method for realizing wear reverse equalization according to claim 1, characterized in that the steps S8 and S9 are both followed by a return to the step S3.
10. The method for realizing wear reverse balance according to claim 1, wherein the member disks of RAID5 corresponding to the method are divided into a plurality of data blocks, one data block is taken from each member disk to form a data stripe, one stripe is composed of N-1 data blocks D and 1 check block P, the data block D is used for storing actual data, and the check block P is used for storing check code data obtained by xor operation of the N-1 data blocks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110640205.6A CN113535082A (en) | 2021-06-09 | 2021-06-09 | Method for realizing wear inverse equilibrium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110640205.6A CN113535082A (en) | 2021-06-09 | 2021-06-09 | Method for realizing wear inverse equilibrium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113535082A true CN113535082A (en) | 2021-10-22 |
Family
ID=78095704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110640205.6A Pending CN113535082A (en) | 2021-06-09 | 2021-06-09 | Method for realizing wear inverse equilibrium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113535082A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114327279A (en) * | 2021-12-29 | 2022-04-12 | 湖南国科微电子股份有限公司 | Data migration method and device |
| CN114546272A (en) * | 2022-02-18 | 2022-05-27 | 山东云海国创云计算装备产业创新中心有限公司 | Method, system, apparatus and storage medium for fast universal RAID demotion to RAID5 |
| CN116774948A (en) * | 2023-08-18 | 2023-09-19 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004185256A (en) * | 2002-12-03 | 2004-07-02 | Hitachi Ltd | Raid system |
| US20060085674A1 (en) * | 2004-10-02 | 2006-04-20 | Hewlett-Packard Development Company, L.P. | Method and system for storing data |
| CN107544747A (en) * | 2016-06-29 | 2018-01-05 | 北京忆恒创源科技有限公司 | Abrasion inverse equalization methods and device |
| CN108334278A (en) * | 2017-12-15 | 2018-07-27 | 中兴通讯股份有限公司 | A kind of storage system balanced management method and apparatus |
| CN109062506A (en) * | 2018-07-16 | 2018-12-21 | 山东大学 | The full flash memory disk array of RAID-5 and the method for improving array bulk life time |
| CN110109621A (en) * | 2019-04-28 | 2019-08-09 | 南威软件股份有限公司 | A method of promoting solid state hard disk array system performance and service life |
-
2021
- 2021-06-09 CN CN202110640205.6A patent/CN113535082A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004185256A (en) * | 2002-12-03 | 2004-07-02 | Hitachi Ltd | Raid system |
| US20060085674A1 (en) * | 2004-10-02 | 2006-04-20 | Hewlett-Packard Development Company, L.P. | Method and system for storing data |
| CN107544747A (en) * | 2016-06-29 | 2018-01-05 | 北京忆恒创源科技有限公司 | Abrasion inverse equalization methods and device |
| CN108334278A (en) * | 2017-12-15 | 2018-07-27 | 中兴通讯股份有限公司 | A kind of storage system balanced management method and apparatus |
| CN109062506A (en) * | 2018-07-16 | 2018-12-21 | 山东大学 | The full flash memory disk array of RAID-5 and the method for improving array bulk life time |
| CN110109621A (en) * | 2019-04-28 | 2019-08-09 | 南威软件股份有限公司 | A method of promoting solid state hard disk array system performance and service life |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114327279A (en) * | 2021-12-29 | 2022-04-12 | 湖南国科微电子股份有限公司 | Data migration method and device |
| CN114546272A (en) * | 2022-02-18 | 2022-05-27 | 山东云海国创云计算装备产业创新中心有限公司 | Method, system, apparatus and storage medium for fast universal RAID demotion to RAID5 |
| CN114546272B (en) * | 2022-02-18 | 2024-04-26 | 山东云海国创云计算装备产业创新中心有限公司 | Method, system, device and storage medium for degrading RAID (redundant array of independent disks) to RAID5 (redundant array of independent disks) |
| CN116774948A (en) * | 2023-08-18 | 2023-09-19 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
| CN116774948B (en) * | 2023-08-18 | 2023-11-28 | 合肥联宝信息技术有限公司 | Data processing method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113535082A (en) | Method for realizing wear inverse equilibrium | |
| US6434666B1 (en) | Memory control apparatus and method for storing data in a selected cache memory based on whether a group or slot number is odd or even | |
| US7809979B2 (en) | Storage control apparatus and method | |
| US9575844B2 (en) | Mass storage device and method of operating the same to back up data stored in volatile memory | |
| US8356292B2 (en) | Method for updating control program of physical storage device in storage virtualization system and storage virtualization controller and system thereof | |
| CN109062506B (en) | RAID-5 full flash disk array and method for prolonging overall service life of array | |
| US20150286531A1 (en) | Raid storage processing | |
| US20140281689A1 (en) | Method and system for rebuilding data following a disk failure within a raid storage system | |
| US20150143021A1 (en) | Equalizing wear on storage devices through file system controls | |
| KR20190052083A (en) | Method and device for writing stored data to a storage medium based on flash memory | |
| US11449402B2 (en) | Handling of offline storage disk | |
| US20050091452A1 (en) | System and method for reducing data loss in disk arrays by establishing data redundancy on demand | |
| CN109725838B (en) | Method, apparatus and computer readable medium for managing a plurality of discs | |
| CN109032513B (en) | RAID (redundant array of independent disks) architecture based on SSD (solid State disk) and HDD (hard disk drive) and backup and reconstruction methods thereof | |
| US20210318739A1 (en) | Systems and methods for managing reduced power failure energy requirements on a solid state drive | |
| US20140181430A1 (en) | Equalizing wear on storage devices through file system controls | |
| CN114489504B (en) | SSD data management method and related components | |
| US9547554B2 (en) | Mass storage device and method of operating the same to store parity data | |
| CN114924691A (en) | Wear leveling method for data stripe reconstruction in RAID5 | |
| US10452291B2 (en) | Control device, storage device and control method | |
| US8140752B2 (en) | Method of executing a background task and an array controller | |
| CN107665090B (en) | Method for realizing global wear leveling | |
| US6754770B2 (en) | Coherence preservation method of duplicated data in RAID subsystem | |
| JP2015179419A (en) | Cache control apparatus, control method thereof, storage device, and computer program | |
| JP4740354B2 (en) | Disk control apparatus and disk control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |