JP6805501B2 - Storage device - Google Patents

Description

The present invention relates to a storage device, and more particularly to a storage device that writes data to a storage medium whose service life is shortened according to the number of writes.

In recent years, flash memory has become widespread due to its large capacity, high recording / playback speed, and low price. Also, in enterprise storage devices, products using only flash memory are being sold.

On the other hand, the problem of flash memory is that the number of writable times is limited. Therefore, wear leveling (wear smoothing), which is a technique for preventing writing from being concentrated on a specific memory cell and extending the service life of the flash memory, is required (for example, paragraph 0039 of Patent Document 1). In particular, when flash memory is used in enterprise products that require high reliability, higher wear leveling is required.

Here, as a wear leveling method, there is a method of equalizing the number of writes based on the statistical information of the number of writes of the memory cell. Specifically, the method of determining the memory cell to write to based on the I / O (Input / Output) statistical information, and the data so that the number of writes between the memory cells is leveled based on the I / O statistical information. There is a way to relocate.

Japanese Unexamined Patent Publication No. 2014-225297

However, the wear leveling method using the above-mentioned I / O statistical information has a problem that it is difficult to extend the service life of the flash memory. That is, with the above-mentioned method, when the behavior of the I / O changes, it is difficult to predict whether the data in the memory cell will be rewritten in the future, and efficient wear leveling processing becomes difficult. As a result, the service life of the flash memory is shortened.

The above-mentioned problems can occur not only in flash memory but also in any storage medium whose service life is shortened by rewriting data.

Therefore, an object of the present invention is to solve the above-mentioned problem that the service life of the storage medium is shortened.

The storage device according to one embodiment of the present invention
A storage device that writes data to a storage medium whose service life is shortened by rewriting.
A shared number acquisition unit that acquires the number of shares representing the number of shared data stored in the storage medium, and a shared number acquisition unit.
A rewriting unit that rewrites the data stored in the storage medium to another storage area having a different service life based on the number of shares.
With,
It takes the configuration.

Moreover, the program which is one form of the present invention
In the control unit equipped in the storage device that writes data to the storage medium whose service life is shortened by rewriting,
A shared number acquisition unit that acquires the number of shares representing the number of shared data stored in the storage medium, and a shared number acquisition unit.
A rewriting unit that rewrites the data stored in the storage medium to another storage area having a different service life based on the number of shares.
To realize,
It takes the configuration.

Further, the rewriting method, which is one form of the present invention, is
This is a rewriting method using a storage device that writes data to a storage medium whose service life is shortened by rewriting.
Acquires the number of shares that represents the number of data stored in the storage medium.
Based on the number of shares, the data stored in the storage medium is rewritten into another storage area having a different service life.
It takes the configuration.

The present invention can suppress the shortening of the service life of the storage medium by being configured as described above.

It is a functional block diagram which shows the structure of the storage device in Embodiment 1 of this invention. It is a figure which shows an example of the structure of the data stored in the storage device disclosed in FIG. It is a figure which shows an example of the data structure of the content address disclosed in FIG. It is a flowchart which shows the operation of the data rewriting process by a storage device. It is a figure which shows the state of the data rewriting process by a storage device. It is a figure which shows the other state of the data rewriting process by a storage device. It is a figure which shows the state of the data rewriting process by the storage apparatus in Embodiment 2 of this invention. It is a functional block diagram which shows the structure of the storage device in Embodiment 3 of this invention.

<Embodiment 1>
The first embodiment of the present invention will be described with reference to FIGS. 1 to 6. 1 to 3 are diagrams for explaining the configuration of the storage device, and FIGS. 4 to 6 are diagrams for explaining the data rewriting process by the storage device.

[Constitution]
The storage device 10 in the present invention is a device that reads and writes data to and from a storage medium. In this embodiment, as shown in FIG. 1, the storage device 10 is connected to the flash memory 20 which is a storage medium. At this time, the storage device 10 and the flash memory 20 are connected by storage interfaces such as SCSI (Small Computer System Interface), SAS (Serial Attached SCSI), SATA (Serial ATA), and FC (Fibre Channel). .. Although the flash memory 20 is described as being arranged outside the storage device 10 in FIG. 1, the flash memory 20 may be mounted inside the storage device 10.

Here, the flash memory 20 includes a plurality of memory cells that are units of a storage area for storing data, and the number of writable times is limited for each memory cell. That is, the flash memory 20 has a property that the service life is shortened by rewriting the data. For this reason, wear leveling is required to prevent writing from being concentrated on a specific memory cell. The present invention is characterized by a wear leveling method, which will be described below. Although the flash memory 20 is described as one in FIG. 1, one or a plurality of flash memories 20 may be used.

The storage device 10 includes a calculation unit and a storage unit (not shown). Then, the calculation unit of the storage device 10 is a data writing unit 11, a duplication number recording unit 12, a duplication number acquisition unit 13, a data rearrangement start unit 14, and a scheduler, which are realized by executing the program by the calculation unit. It is equipped with 15.

Further, the flash memory 20 includes an arithmetic unit and a storage unit (not shown). The arithmetic unit of the flash memory 20 is a data writing unit 21, a writing count recording processing unit 22, a writing count acquisition unit 23, a data rearrangement unit 24, and an address, which are realized by executing the program by the arithmetic unit. The conversion unit 25 is provided. The storage unit of the flash memory 20 includes an address translation recording unit 26, a write count recording unit 27, and a data storage unit 28.

Hereinafter, each configuration included in the storage device 10 and the flash memory 20 will be described in detail.

The data writing unit 11 of the storage device 10 writes data in response to a request from the user. For example, as the data writing unit 11, a protocol such as CIFS / NFS (Common Internet File System / Network File System) can be mentioned. Specifically, the data writing unit 11 of the storage device 10 writes data in cooperation with the data writing unit 21 of the flash memory 20. Here, the state at the time of writing data will be described with reference to FIGS. 2 and 3.

The data writing units 11 and 21 first divide the file to be written into block data having a predetermined capacity, and store the block data (actual data) in the memory cell serving as the data storage unit 28 of the flash memory 20. .. At this time, the data writing units 11 and 12 generate address data (content address: CA) based on the data content and storage position of the block data, which refers to the block data, and flash the block data together with the content address in a tree structure. Store in memory 20. Therefore, even if the block data of different files are the same, the data writing units 11 and 21 may refer to the same address data as long as the data contents are the same, so that one block data is stored in the storage device. It will be good to do it. As described above, the data writing units 11 and 21 have a function of eliminating duplicate storage of block data having the same content (duplicate storage exclusion function).

As an example, in the example of FIG. 2, for example, it is assumed that the file 1 is composed of three block data 11, 12, and 13, and these are stored in the memory cell. Then, the file 1 can be read by tracing the content addresses CA11,12,13 that refer to the storage positions of the block data 11,12,13. On the other hand, it is assumed that the file 2 is composed of three block data 21, 22, 23, and these are stored in the memory cell. When the block data 21 has the same contents as the block data 13 constituting the file 1, the content address CA13 of the block data 13 already stored is referred to without storing the actual data of the block data 21. You just have to do it. As a result, the already stored block data 13 is stored as the block data of the file 2.

Then, the duplicate count recording unit 12 of the storage device 10 stores the number of references (number of shares) indicating how many files the block data is referenced and shared for each block data stored as described above. To do. For example, as shown in FIG. 3, information that identifies a file that refers to the block data, for example, an i-node number, is recorded in the content address CA that refers to each block data. That is, the number of i-node numbers in the content address CA that refers to the block data is how many files refer to the block data referenced by the content address CA, that is, how many files share it. It is the number of references (number of shares) that represents. For example, in the example of FIG. 2, since the block data 13 is referenced (shared) by the file 1 and the file 2, the number of references to the block data 13 is “2”.

Further, when writing block data, the address conversion unit 25 of the flash memory 20 stores information indicating the correspondence between the position where the block data is written and the memory cell in which the block data is actually stored, in the address conversion recording unit. Record at 26. By this conversion, as will be described later, even if the arrangement position of the block data on the memory cell is changed, the block data can be accessed from the storage device 10 by designating the same writing position. In this way, the address translation unit 25 updates the correspondence between the data arrangement position visible from the storage device 10 and the data arrangement position on the memory cell.

Further, the write count recording processing unit 22 of the flash memory 20 counts the write count for each memory cell and records it in the write count recording unit 27. For example, when the memory cell is updated by the data writing unit 21 or the like, the write count recording processing unit 22 increases the write count of the memory cell by one and records it in the write count recording unit 27. The number of writes for each memory cell is also counted when the block data is rewritten, which will be described later.

Next, a configuration for relocating the block data by the storage device 10 will be described. First, the scheduler 15 of the storage device 10 periodically calls and executes the data rearrangement start unit 14. In response to this, the data rearrangement start unit 14 executes wear leveling, which is a process of rewriting block data (actual data) stored in each memory cell in the flash memory 20.

In wear leveling, the following processing is executed. First, the duplicate count acquisition unit 13 (shared number acquisition unit) of the storage device 10 refers to each block data from a number of files based on the content address CA that refers to each block data stored in the flash memory 20. Get the number of references that indicate whether or not. Specifically, as shown in FIG. 2, the number of i-node numbers stored in the content address CA that refers to the block data is acquired as the number of references to the block data.

Further, the write count acquisition unit 23 of the flash memory 20 acquires the write count of the memory cell in which each block data is arranged from the write count recording unit 27.

Then, the data relocation unit 24 (rewriting unit) receives the position, the number of times the memory cell is written, and the number of references to the block data stored in the memory cell for each memory cell. Then, the data rearrangement unit 24 actually performs wear leveling, that is, rewriting the block data stored in the memory cell, based on the information. This rewriting process is, for example, a process of rewriting block data having a large number of references to another memory cell having a short usage life, that is, another memory cell having a large number of writes. A specific example of the rewriting process will be described in detail when the operation is described below.

Further, the address translation unit 25 described above records information indicating the correspondence between the position where the block data is stored and the actual memory cell even when the block data is rewritten by the data rearrangement unit 24. Record in section 26.

[motion]
Next, the operation of the storage device 10 having the above-described configuration, particularly the wear leveling operation, will be described with reference to the flowchart of FIG. 4 and the diagrams showing the state of the block data of FIGS. 5 and 6.

First, the scheduler 15 periodically calls the data rearrangement start unit 14. The data relocation start unit 14 first acquires the write count of each memory cell acquired by the write count acquisition unit 23, and refers to the block data stored in each memory cell acquired by the duplicate count acquisition unit 13. Get the number.

Then, the data relocation start unit 14 confirms from the acquired number of writes and the number of references whether there is block data having a certain number of references or more in the memory cell having a number of writes less than a certain number (step S1). ). If such block data does not exist (No in step S1), wear leveling according to the present invention ends without being carried out. If such block data exists (Yes in step S1), the block data is selected as a replacement candidate (rewrite candidate) in the subsequent steps (step S2).

Whether or not the number of writes in the memory cell is less than a certain number is determined by whether or not the number of writes in the memory cell is smaller than the preset write count threshold value. Further, whether or not the number of references to the block data is equal to or greater than a certain number is determined by whether or not the number of references to the block data is equal to or greater than a preset reference number threshold value.

Subsequently, the data relocation start unit 14 confirms whether or not there is a free memory cell in which no other data is stored among the memory cells whose number of writes is a certain number or more (step S3). Whether or not the number of writes in the memory cell is equal to or greater than a certain number is determined by whether or not the number of writes in the memory cell is equal to or greater than the above-mentioned write count threshold value. It may be performed depending on whether or not it is equal to or higher than the threshold value.

Then, when there are free memory cells whose number of writes is a certain number or more (Yes in step S3), the data rearrangement unit 24 moves the block data selected as the replacement candidate in step S2 to the free memory cells. (Step S4). That is, as shown in FIG. 5A, the data rearrangement unit 24 displays block data stored in a memory cell having a small number of writes less than a certain number and having a large number of references of a certain number or more in FIG. As shown in (B), the data is stored again in the memory cell having a large number of writes of a certain number or more.

If there is no free memory cell with a certain number of writes or more in step S3 (No in step S3), the data relocation start unit 14 refers to a memory cell with a certain number of writes or more with less than a certain number. It is confirmed whether there is block data having a number (step S5). Whether or not the number of references to the block data is less than a certain number is determined by whether or not the number of references to the block data is less than the above-mentioned reference number threshold value, but other values smaller than the reference number threshold value are used. This may be done depending on whether or not it is below the threshold value. If there is no such block data (No in step S5), it is determined that there is no destination for the block data selected as the replacement candidate in step S2, and wear leveling is not performed.

On the other hand, in step S5, if there is block data having a reference number less than a certain number in the memory cell whose number of writes is a certain number or more (Yes in step S5), the data rearrangement unit 24 is replaced in step S2. The arrangement of the block data selected as the candidate and the block data found in step S5 on the memory cell is exchanged (step S6). That is, as shown in FIG. 6A, the data relocation unit 24 displays block data having a large number of references of a certain number or more stored in the memory cell, which is a small number of writes less than a certain number, in FIG. As shown in (B), the data is stored again in the memory cell having a large number of writes of a certain number or more. At the same time, as shown in FIG. 6A, block data having a small number of references less than a certain number stored in a memory cell having a large number of writes of a certain number or more is shown in FIG. 6B. As described above, the data is re-stored in a memory cell having a small number of writes less than a certain number.

In step S5, even if there is no block data having a reference number less than a certain number in the memory cells whose number of writes is a certain number or more (No in step S5), the data rearrangement start unit 14 still performs. Wear leveling may be performed as follows. First, the number of references to block data stored in a memory cell whose number of writes is a certain number or more is compared with the number of references to block data selected as replacement candidates in step S2. Then, as a result of the comparison, the block data having a large number of references is rewritten so as to be arranged in a memory cell having a certain number of writes or more. For example, when the number of references to the block data selected as the replacement candidate in step S2 is larger, the block data is rewritten into a memory cell having a certain number of writes or more. At the same time, the block data stored in the memory cell having a certain number of writes or more is rewritten to the memory cell having a small number of writes less than a certain number.

Then, after rewriting the block data by processing such as step S4 and step S6, the address conversion unit 25 updates the address conversion recording unit 26 so that the position of each block data corresponds to the position on the memory cell. To do.

As described above, in the present invention, the block data having a large number of references is rewritten so as to be arranged in the memory cell having a large number of writes. Here, since the block data having a large number of references is referenced from various files, it is not released unless all the referenced files are deleted. Therefore, the block data having a large number of references is unlikely to be deleted thereafter, and the memory cell in which the block data is stored is unlikely to be rewritten. On the contrary, since the block data having a small number of references is likely to be deleted later, the memory cell in which the block data is stored is likely to be rewritten.

For this reason, as described above, by arranging block data that is unlikely to be rewritten in the future in a memory cell having a large number of writes, it is possible to suppress an increase in the number of subsequent writes to the memory cell having a large number of writes. can do. As a result, it is possible to extend the life of the flash memory.

In the above, the block data in which the number of writes is less than a certain number and the number of references is a certain number or more is rewritten into a memory cell in which the number of writes is a certain number or more. However, the present invention rewrites by such a method. Not limited to that. For example, the block data may be stored in another memory cell according to the number of references to the block data. At this time, as the other memory cell to be rewritten, a memory cell having a different service life may be selected from the number of writes as described above, or a memory cell having a different service life may be selected by another method.

Further, in the above, as shown in FIG. 3, the i-node number of the file that refers to the block is recorded for each content address, and the number of references is acquired. However, the number of references may be obtained by other methods without recording such persistent information. For example, in a storage device that performs deduplication, block data is not released until the number of references reaches 0, and garbage collection is performed periodically to release the block whose number of references becomes 0. .. In this garbage collection, it is also possible to confirm which block data the i-node refers to and create the content shown in FIG. Then, after that, the same wear leveling as described above can be performed by calling the data rearrangement start unit 14. According to this method, the contents of FIG. 3 can be released after the processing of the data rearrangement start unit 14 is completed.

Further, in the above, wear leveling is performed based on the number of reference (multiple) of block data in the storage device having the deduplication function, but the configuration of the present invention can also be used in other storage devices. .. For example, in a file system or a RAW device having a snapshot function, data is shared between snapshots, and wear leveling may be performed using the number of shared data as the above-mentioned reference number.

<Embodiment 2>
Next, a second embodiment of the present invention will be described with reference to FIG. FIG. 7 is a diagram for explaining a data rewriting process by the storage device.

The storage device 10 in the present invention has substantially the same configuration as that of the first embodiment described above, but the configuration of the flash memory 20 is different. The flash memory 20 in the present embodiment contains a mixture of SLC (Single Level Cell), MLC (Multi Level Cell), and TLC (Triple Level Cell). In such a configuration, the data rearrangement unit 24 of the storage device 10 has a function of selecting an appropriate flash memory as described below according to the number of reference to the block data.

Here, SLC, MLC, and TLC have different performances and prices, and in particular, the service life, that is, the number of rewritable times is set. For example, MLC and TLC are set to have a smaller number of rewritable times than SLC, and have a short service life.

Then, as shown in FIG. 7A, for example, the data rearrangement unit 24 displays the block data stored in the SLC having a large number of rewritable times and having a large number of references of a certain number or more in FIG. 7B. As shown in, rewrite to MLC with a small number of rewritable times.

In this way, by arranging the block data having a large number of references and having a low possibility of being deleted in the MLC having a small number of rewritable times, it is possible to suppress an increase in the number of subsequent writes to the MLC. As a result, even a flash memory having a short service life can be extended in life.

<Embodiment 3>
Next, a third embodiment of the present invention will be described with reference to FIG. FIG. 8 is a block diagram showing the configuration of the storage device.

The storage device 100 in the present invention is a storage device 100 that writes data to a storage medium 110 whose service life is shortened by rewriting, and is a shared number representing the number of data stored in the storage medium 110. A shared number acquisition unit 101 for acquiring the data, and a rewriting unit 102 for rewriting the data stored in the storage medium 110 to another storage area having a different usage life based on the shared number. The shared number acquisition unit 101 and the rewrite unit 102 are constructed by executing a program in a control unit (not shown) provided in the storage device 100.

Then, in the storage device 100, first, the shared number acquisition unit 101 acquires the shared number representing the number of data stored in the storage medium 110 being shared. Then, the rewriting unit 102 rewrites the data stored in the storage medium 110 to another storage area having a different service life based on the acquired number of shares.

In this way, by rewriting the data to another storage area having a different service life based on the number of shared data, the number of writes to the storage medium can be suppressed, and the life of the storage medium can be extended. Can be done.

<Additional notes>
Part or all of the above embodiments may also be described as in the appendix below. Hereinafter, the outline of the configuration of the storage device, the program, and the rewriting method in the present invention will be described. However, the present invention is not limited to the following configurations.

(Appendix 1)
A storage device that writes data to a storage medium whose service life is shortened by rewriting.
A shared number acquisition unit that acquires the number of shares representing the number of shared data stored in the storage medium, and a shared number acquisition unit.
A rewriting unit that rewrites the data stored in the storage medium to another storage area having a different service life based on the number of shares.
Storage device with.

(Appendix 2)
The storage device according to Appendix 1.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage area having a shorter service life than the storage area in which the data is stored.
Storage device.

(Appendix 3)
The storage device according to Appendix 1 or 2.
A write count acquisition unit for acquiring the write count for each storage area formed in the storage medium is provided.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage area having a larger number of writes than the storage area in which the data is stored.
Storage device.

(Appendix 4)
The storage device according to Appendix 3,
The rewriting unit is stored in a storage area that is less than the preset number of writes, and rewrites data that is equal to or greater than the preset number of shares to another storage area that is greater than or equal to the preset number of writes.
Storage device.

(Appendix 5)
The storage device according to Appendix 4,
The rewriting unit is stored in a storage area that is less than the preset number of writes, and stores data that is greater than or equal to the preset number of shares among other storage areas that are greater than or equal to the preset number of writes. Rewrite to a free storage area where the data of
Storage device.

(Appendix 6)
The storage device according to Appendix 4 or 5.
The rewriting unit is stored in a storage area less than the preset number of writes, and stores data having a number of shares less than the preset number of shares, and data having a number of shares less than the number of shares of the data. Rewrite to another storage area that is equal to or greater than the preset number of writes.
Storage device.

(Appendix 7)
The storage device according to Appendix 2,
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage medium having a shorter service life than the storage medium in which the data is stored.
Storage device.

(Appendix 8)
The storage device according to Appendix 7.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage medium set to have a smaller number of rewritable times than the storage medium in which the data is stored.
Storage device.

(Appendix 9)
In the control unit equipped in the storage device that writes data to the storage medium whose service life is shortened by rewriting,
A shared number acquisition unit that acquires the number of shares representing the number of shared data stored in the storage medium, and a shared number acquisition unit.
A rewriting unit that rewrites the data stored in the storage medium to another storage area having a different service life based on the number of shares.
A program to realize.

(Appendix 9-1)
The program described in Appendix 9
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage area having a shorter service life than the storage area in which the data is stored.
program.

(Appendix 9-2)
The program in Appendix 9 or 9-1
The control unit further realizes a write count acquisition unit that acquires the write count for each storage area formed in the storage medium.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage area having a larger number of writes than the storage area in which the data is stored.
program.

(Appendix 9-3)
The program described in Appendix 9-1.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage medium having a shorter service life than the storage medium in which the data is stored.
program.

(Appendix 9-4)
The program described in Appendix 9-3.
The rewriting unit rewrites the data determined to have a large number of shares according to a preset standard to another storage medium set to have a smaller number of rewritable times than the storage medium in which the data is stored.
program.

(Appendix 10)
This is a rewriting method using a storage device that writes data to a storage medium whose service life is shortened by rewriting.
Acquires the number of shares that represents the number of data stored in the storage medium.
Based on the number of shares, the data stored in the storage medium is rewritten into another storage area having a different service life.
How to rewrite.

(Appendix 10-1)
The rewriting method described in Appendix 10
The data determined to have a large number of shares according to a preset standard is rewritten to another storage area having a shorter service life than the storage area in which the data is stored.
How to rewrite.

(Appendix 10-2)
It is a rewriting method to Appendix 10 or 10-1.
Further, the number of writes for each storage area formed in the storage medium is acquired, and the number of writes is acquired.
The data determined to have a large number of shares according to a preset standard is rewritten to another storage area having a larger number of writes than the storage area in which the data is stored.
How to rewrite.

(Appendix 10-3)
The rewriting method described in Appendix 10-1.
The data determined to have a large number of shares according to a preset standard is rewritten to another storage medium having a shorter service life than the storage medium in which the data is stored.
How to rewrite.

(Appendix 10-4)
The rewriting method described in Appendix 10-3.
The data determined to have a large number of shares according to a preset standard is rewritten to another storage medium set to have a smaller number of rewritable times than the storage medium in which the data is stored.
How to rewrite.

The above-mentioned program is stored in a storage device or recorded on a computer-readable recording medium. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

Although the invention of the present application has been described above with reference to the above-described embodiments and the like, the present invention is not limited to the above-described embodiments. Various changes that can be understood by those skilled in the art can be made to the structure and details of the present invention within the scope of the present invention.

10 Storage device 11 Data write unit 12 Duplicate count recording unit 13 Duplicate count acquisition unit 14 Data relocation start unit 15 Scheduler 20 Flash memory 21 Data write unit 22 Write count recording processing unit 23 Write count acquisition unit 24 Data relocation unit 25 Address Conversion unit 26 Address conversion acquisition unit 27 Write count recording unit 28 Data storage unit 100 Storage device 101 Shared number acquisition unit 102 Rewriting unit 110 Storage medium

Claims (7)

A storage device that writes data to a storage medium whose service life is shortened by rewriting.
When the file to be written is divided into block data and the block data is written to the storage medium, the data content of the block data and the address data based on the storage position that refer to the block data are generated, and the data content is A data writing unit having a function of eliminating duplicate storage of the block data having the same contents by referring to the same block data with the same address data.
In the address data that refers to the block data, a duplicate number recording unit that stores the number of the files in which the block data is shared in the file, and
Shared number the block data stored in the storage medium represents the number of the files shared in the file, a shared number obtaining unit for obtaining on the basis of the address data that refers to the block data,
A write count acquisition unit that acquires the write count for each storage area formed on the storage medium,
A rewriting unit that rewrites the block data, which is determined to have a large number of shares according to a preset standard, to another storage area that has a larger number of writes than the storage area in which the block data is stored.
Storage device with. The storage device according to claim 1.
The rewriting unit is stored in a storage area that is less than the preset number of writes, and stores the block data that is equal to or greater than the preset number of shares in another storage area that is greater than or equal to the preset number of writes. rewrite,
Storage device. The storage device according to claim 2.
The rewriting unit is stored in a storage area having a preset number of writes or less, and the block data having a preset number of shares or more can be stored in a storage area having a preset number of writes or more. Rewrite to a free storage area where other data is not stored,
Storage device. The storage device according to claim 2 or 3.
The rewriter is stored in the preset memory area of less than the write count was, the block data of the above shared number set in advance, the number of shared shares on the less than of the block data Rewrite to another storage area in which the block data is stored, which is equal to or greater than the preset number of writes.
Storage device. The storage device according to any one of claims 1 to 4.
The rewriting unit transfers the block data, which is determined to have a large number of shares according to a preset standard, to another storage medium set to be rewritable less than the storage medium in which the block data is stored. rewrite,
Storage device. A computer equipped with a storage device that writes data to a storage medium whose service life is shortened by rewriting.
When the file to be written is divided into block data and the block data is written to the storage medium, the data content of the block data and the address data based on the storage position that refer to the block data are generated, and the data content is A data writing unit having a function of eliminating duplicate storage of the block data having the same contents by referring to the same block data with the same address data.
In the address data that refers to the block data, a duplicate number recording unit that stores the number of the files in which the block data is shared in the file, and
Shared number the block data stored in the storage medium represents the number of the files shared in the file, a shared number obtaining unit for obtaining on the basis of the address data that refers to the block data,
A write count acquisition unit that acquires the write count for each storage area formed on the storage medium,
A rewriting unit that rewrites the block data, which is determined to have a large number of shares according to a preset standard, to another storage area that has a larger number of writes than the storage area in which the block data is stored.
A program to realize. This is a rewriting method using a storage device that writes data to a storage medium whose service life is shortened by rewriting.
When the file to be written is divided into block data and the block data is written to the storage medium, the data content of the block data and the address data based on the storage position that refer to the block data are generated, and the data content is By referencing the same block data with the same address data, duplicate storage of the block data having the same content can be eliminated.
In the address data that refers to the block data, the number of the files in which the block data is shared in the file is stored.
Shared number the block data stored in the storage medium represents the number of the files shared in the file, obtains based on the address data that refers to the data of the file,
Acquires the number of writes for each storage area formed in the storage medium,
The block data determined to have a large number of shares according to a preset standard is rewritten to another storage area having a larger number of writes than the storage area in which the block data is stored.
How to rewrite.

Images