JP6842064B2 - Storage device, program, information processing method - Google Patents

Description

The present invention relates to a storage device, a program, and an information processing method, and more particularly to a storage device, a program, and an information processing method that eliminates duplicate storage of data having the same contents.

As a technique for efficiently handling a huge amount of data, a storage device having a function of eliminating duplicate storage is known.

In the case of a storage system that performs deduplication as described above, new data will be added to the end of the storage area, for example. Therefore, when reading data later, it may be necessary to perform a huge number of disk operations in order to read the data spread over the entire storage device.

As a technique for dealing with the above problem, for example, there is Patent Document 1. Patent Document 1 describes a storage device including a data dividing means, a block detecting means, and a data writing means. According to Patent Document 1, the block detecting means includes a plurality of continuous data blocks constituting a predetermined range in the data to be written among the divided data blocks, and a predetermined range already continuously stored in the storage device. Detects a common ratio that represents the ratio of the common part of multiple data blocks. Further, the data writing means newly stores the divided data blocks in the storage device according to the common ratio detected by the block detecting means. According to Patent Document 1, with such a configuration, it is possible to control to write a new data block to the storage device only when the common ratio is smaller than, for example, a predetermined threshold value. As a result, according to Patent Document 1, it is possible to suppress the diffusion of the data block over the entire storage area in the storage device while performing deduplication. This makes it possible to suppress a decrease in reading performance.

Japanese Patent Application Laid-Open No. 2013-541555

In order to read data at high speed, in a storage device or the like having a function of eliminating duplicate storage as described in Patent Document 1, the data pre-read from a hard disk or the like is stored in a storage device such as a buffer and buffered. The data stored in may be used. However, the amount of data that can be stored in the buffer is limited. Therefore, for example, the data stored in the buffer is discarded at the timing when the generation to which the data to be read belongs changes. As a result, when the data once read into the buffer is needed again, there is a problem that the data has already been expelled from the buffer and it may be necessary to read the data from the disk again.

As described above, in the storage device having a function of eliminating duplicate storage, there has been a problem that it is difficult to efficiently acquire necessary data and it is difficult to suppress a decrease in read performance.

Therefore, an object of the present invention is that in a storage device having a function of eliminating duplicate storage, which is the above-mentioned problem, it is difficult to efficiently acquire necessary data, and it is difficult to suppress a deterioration in read performance. The purpose is to provide a storage device, a program, and an information processing method that can solve the problem of.

A storage device according to an embodiment of the present invention for achieving such an object
A data storage unit that deduplications and stores data of multiple generations,
A temporary data storage unit that temporarily stores the data read from the data storage unit, and
A data read control unit that reads data stored in the data storage unit and stores the data in the temporary data storage unit, and also reads data from the temporary data storage unit.
Have,
The data read control unit controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. Take the configuration of doing.

In addition, the program which is another form of the present invention
A storage device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
A data read control means for reading data stored in the data storage unit and storing the data in the temporary data storage unit and reading data from the temporary data storage unit is realized.
The data read control means controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. It is a program to realize the processing to be performed.

Further, the information processing method, which is another embodiment of the present invention, is
An information processing device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
The data stored in the data storage unit is read and stored in the temporary data storage unit, and the data is read from the temporary data storage unit, and the data of each generation among the data read in response to the read request received from the external device. The data storage status for each generation to be stored in the temporary data storage unit is controlled based on the ratio of the data.

The present invention can solve the problem that it is difficult to suppress a decrease in read performance in a storage device having a function of eliminating duplicate storage by being configured as described above. It becomes possible to provide an information processing method.

It is a block diagram which shows an example of the structure of the whole system. It is a block diagram which shows the outline of the structure of the storage system shown in FIG. It is a functional block diagram which shows an example of the structure of the storage system shown in FIG. It is a figure for demonstrating an example of the data writing process in the storage system shown in FIG. It is a figure for demonstrating an example of the data writing process in the storage system shown in FIG. It is a figure which shows an example of how the data reading means shown in FIG. 3 pre-reads. It is a flowchart which shows an example of the data read processing by the storage system in 1st Embodiment of this invention. It is a flowchart which shows a detailed example of the process of reading the fragment shown in FIG. 7. It is a block diagram which shows an example of the structure of the storage device which concerns on 2nd Embodiment of this invention.

[First Embodiment]
The first embodiment of the present invention will be described with reference to FIGS. 1 to 8. FIG. 1 is a block diagram showing an example of the configuration of the entire system. FIG. 2 is a block diagram showing an outline of the configuration of the storage system 1. FIG. 3 is a functional block diagram showing an example of the configuration of the storage system 1. 4 and 5 are diagrams for explaining an example of data writing processing in the storage system 1. FIG. 6 is a diagram showing an example of how the data reading means pre-reads. 7 and 8 are flowcharts showing an example of data read processing by the storage system 1.

In the first embodiment, a storage system 1 having a deduplication function, which suppresses a decrease in read performance by efficiently using the prefetch buffer 16, will be described. Upon receiving the read request, the storage system 1 in the present embodiment restores a series of data blocks and concatenates the restored data blocks to restore the target file. At this time, the storage system 1 acquires a plurality of fragment data generated based on the data block from the disk device 15, and restores the data block based on the acquired fragment data. When acquiring fragment data from the disk device 15, the storage system 1 in the present embodiment acquires necessary fragment data by pre-reading from the disk device 15 and utilizing the fragment data stored in the prefetch buffer 16. To do. Further, as will be described later, the storage system 1 stores the information stored in the prefetch buffer 16 and the stored information based on the ratio of the data for each generation to the data read (read) in response to the read request. Take control. In this way, the pre-read data is effectively utilized by controlling the information stored in the prefetch buffer 16 and the stored information based on the ratio of the data for each generation to the data read in response to the read request. It becomes possible to do. As a result, it is possible to suppress a decrease in read performance in a storage device having a function of eliminating duplicate storage.

Referring to FIG. 1, the storage system 1 in the present embodiment is connected to the backup system 4 that controls the backup process via the network N. Then, the backup system 4 acquires the backup target data (data to be written) stored in the backup target device 5 connected via the network N, and requests the storage system 1 to store the data. As a result, the storage system 1 stores the backup target data requested to be stored for backup.

As shown in FIG. 2, the storage system 1 in the present embodiment has a configuration in which a plurality of server computers are connected. Specifically, the storage system 1 includes an accelerator node 2 which is a server computer that controls a storage / playback operation in the storage system 1 itself, and a storage node 3 which is a server computer provided with a storage device for storing data. There is. The number of accelerator nodes 2 and the number of storage nodes 3 are not limited to those shown in FIG. The number of accelerator nodes 2 and storage nodes 3 may be one, or may be any number of two or more.

Further, the storage system 1 in the present embodiment divides and redundantly divides the data, distributes the data, and stores the data in a plurality of storage devices. The process when the storage system 1 stores data will be described later.

In the following, assuming that the storage system 1 is one system, the configuration and functions provided by the storage system 1 will be described. That is, the configuration and functions of the storage system 1 described below may be provided in either the accelerator node 2 or the storage node 3. As shown in FIG. 2, the storage system 1 is not necessarily limited to including the accelerator node 2 and the storage node 3, and may have any configuration, for example, one computer. It may have been done.

FIG. 3 shows an example of the configuration of the storage system 1 in this embodiment. Referring to FIG. 3, the storage system 1 stores data, a data writing means 11, a data reading means 12 (a part of the data reading control unit), a duplication rate calculating means 13 (a part of the data reading control unit), and data. It has an area control means 14 (a part of a data read control unit), a disk device 15 (data storage unit), and a prefetch buffer 16 (temporary data storage unit). The storage system 1 has, for example, an arithmetic unit (not shown) and a storage device (not shown) that perform predetermined arithmetic processing, and the arithmetic unit executes a program stored in the storage device to realize each of the above means. ..

Actually, the configuration provided in the storage system 1 described above is used in a computing device such as a CPU (Central Processing Unit) and a storage device such as a hard disk provided in the accelerator node 2 and the storage node 3 shown in FIG. It is composed of.

The data writing means 11 stores the deduplicated data in the disk device 15 in response to a file write request.

For example, referring to FIG. 4, the storage system 1 receives the input of the write-requested file A. Then, the data writing means 11 divides the file A into a data block D having a predetermined capacity (for example, 64 KB). By dividing the file A into the data blocks D in this way, deduplication can be easily performed.

Subsequently, the data writing means 11 confirms whether or not the divided data block D has already been written to the storage system 1 (disk device 15). For example, the data writing means 11 uses a preset hash function to calculate a unique hash value representing the data content based on the data content of the data block D. After that, the data writing means 11 uses the calculated hash value to check whether or not the data block having the hash value is already stored in the disk device 15. For example, in this way, the data writing means 11 confirms the presence or absence of duplication depending on whether or not the hash value based on the data content of the data block D is already registered in the storage system 1.

When the storage system 1 already has the calculated hash value, it can be determined that the data block D having the same contents is already stored. In this case, the data writing means 11 specifies the area of the data block D having the same contents already stored as the storage destination of the data block D for which the writing is requested, so that the data block D for which the writing is requested is stored. Is memorized. This eliminates the need to actually store the data block D related to the write request in the storage device 20. In other words, when the data block D having the same contents has already been written, the data writing means 11 does not write the fragment data described later.

On the other hand, when it is determined that the data block D related to the write request is not yet stored, the data writing means 11 compresses the data block D related to the write request and divides the data block D related to the write request into a plurality of fragment data having a predetermined capacity. .. For example, the data writing means 11 divides into nine fragment data as shown by reference numerals D1 to D9 in FIG. Further, the data writing means 11 generates redundant data so that the original data block can be restored even if some of the divided fragment data is missing. Then, the data writing means 11 adds the generated redundant data to the divided fragment data. For example, the data writing means 11 adds three fragment data (redundant data) as shown by reference numerals D10 to D12 in FIG. As a result, the data writing means 11 generates a data set consisting of 12 fragment data composed of 9 fragment data and 3 redundant data.

Subsequently, the data writing means 11 distributes and stores each fragment data constituting the data set generated as described above in each storage area formed in the storage device (disk device 15). For example, the data writing means 11 calculates a component to which the fragment data belongs based on the fragment data, and stores the fragment data in the calculated component. The component refers to a logical concept possessed by each storage node 3 (storage device), and fragment data is stored in the component. Further, in the present embodiment, the calculation method of the component to which the fragment data belongs is not particularly limited. The data writing means 11 can calculate the component to which the fragment data belongs by using various known means.

In addition, the data writing means 11 writes the fragment data and acquires management information including control information such as component configuration, fragment data storage file (generation), position indicating information, and redundant information. Then, when the data writing means 11 stores the fragment data in the component, the data writing means 11 also stores the management information in the disk device 15 in which the component is arranged.

The data writing means 11 stores the deduplicated data in the disk device 15 as described above, for example. As a result, for example, as shown in FIG. 5, when a new generation file is to be saved, only the data (difference data) added / updated in the new generation is stored in the disk device 15 which is a storage device. Will be done.

The data reading means 12 reads the data stored in a state of being deduplicated by the data writing means 11 in response to a read request received from the backup system 4 or the like which is an external device.

For example, when the data reading means 12 receives the reading request, the data reading means 12 reads the necessary fragment data from the disk device 15 or the prefetch buffer 16 with reference to the management information, and restores the data block D. Then, the data reading means 12 concatenates a plurality of data blocks D restored by repeating the above means, and restores a group of data such as a file.

For example, the data reading means 12 reads the fragment data from the prefetch buffer 16 when the fragment data to be read is stored in the prefetch buffer 16 by the look-ahead described later. On the other hand, when the fragment data to be read is not stored in the prefetch buffer 16, the data reading means 12 reads the fragment data from the disk device 15. In this way, the data reading means 12 reads the necessary fragment data from the disk device 15 or the prefetch buffer 16. Then, the data reading means 12 restores the data block D from each fragment data read in response to the reading request. Further, the data reading means 12 concatenates a plurality of data blocks D restored by repeating the above means, and restores a group of data such as a file.

Further, the data reading means 12 in the present embodiment determines whether or not to pre-read the fragment data based on the duplication rate calculated by the duplication rate calculating means 13 described later when reading the fragment data from the disk device 15. (See FIG. 6). For example, in the data reading means 12, when the duplication rate of the generation to which the fragment data to be read belongs is lower than a predetermined reference value (any value may be used, for example, 0.3) (look-ahead reference value). , Do not look ahead. On the other hand, when the duplication rate of the generation to which the fragment data to be read belongs is equal to or higher than the look-ahead reference value, the data reading means 12 has a predetermined number (arbitrary number may be any number) of the fragment data to be read continuous on the disk device 15. No, for example, 4) subsequent fragment data is pre-read, for example, in the background. For example, in the case shown in FIG. 6, the data reading means 12 reads the fragment data A to be read from the disk device 15 and pre-reads the subsequent fragment data B, C, D, and E in the background. In this way, the data reading means 12 acquires the fragment data to be read from the disk device 15 based on the duplication rate, and pre-reads a predetermined number (predetermined range) of fragment data following the read target. Then, the data reading means 12 stores a predetermined number of pre-read subsequent fragment data in the data storage area 161 of the corresponding generation of the prefetch buffer 16.

In this way, the data reading means 12 reads the data stored by the data writing means 11 in response to the reading request.

The duplication rate calculating means 13 calculates the duplication rate, which is the ratio of the data of each generation to the (read) data read in response to the read request, for example, in response to an instruction from the data reading means 12. .. Then, the duplication rate calculating means 13 stores the calculated duplication rate in a storage device such as a memory (not shown).

For example, the duplication rate calculating means 13 divides the data size of the generation n read in response to the read request by the total data size read in response to the read request. As a result, the duplication rate calculation means 13 calculates the duplication rate of the generation n. Note that n may be any value.

The duplication rate calculating means 13 is, for example, the timing at which the generation of the fragment data read by the data reading means 12 changes (the generation to which the fragment data belongs is determined, for example, according to the storage position of the fragment data), and the prefetch buffer described later. When the number of data storage areas 161 secured in 16 exceeds a predetermined number, the required duplication rate of generations is calculated. For example, the duplication rate calculating means 13 has a duplication rate of the generation before the change (before the change) (the generation to which the fragment immediately before the fragment whose generation changes) belongs at the timing when the generation of the fragment data read by the data reading means 12 changes. Is calculated. Further, the duplication rate calculation means 13 calculates the duplication rate of each generation in which the data storage area 161 is secured at the timing when the number of data storage areas 161 secured in the prefetch buffer 16 exceeds a predetermined number. Further, the duplication rate calculating means 13 calculates the duplication rate of the generation to which the fragment data to be read belongs at the timing of reading the fragment data from the disk device 15. The duplication rate calculation means 13 may be configured to calculate the duplication rate at a timing other than the above timing (for example, periodically).

The data storage area control means 14 controls the prefetch buffer 16 based on the instruction from the data reading means 12 and the duplication rate calculated by the duplication rate calculation means 13.

For example, when the generation of the fragment data read by the data reading means 12 changes, the data storage area control means 14 changes the generation (generation changes) after the change (after the change) based on the instruction from the data reading means 12. The data storage area 161 for the generation to which the data belongs is secured in the prefetch buffer 16. If the data storage area 161 for the corresponding generation is already secured in the prefetch buffer 16, the data storage area control means 14 does not have to secure the data storage area 161 for the generation. Further, the data storage area control means 14 is for the generation before the change based on the overlap rate of the generation before the change (the generation to which the fragment immediately before the fragment whose generation changes) calculated by the duplication rate calculation means 13. It is determined whether or not to discard the data storage area 161. For example, in the data storage area control means 14, when the duplication rate of the generation before the change is lower than a predetermined reference value (optional, for example, 0.3) (discard reference value), the generation before the change. Data storage area 161 for use is discarded. When the data storage area control means 14 discards the data storage area 161, the fragment data stored in the data storage area 161 (fragment data of the generation corresponding to the data storage area 161 to be discarded) is also discarded. ..

Further, the data storage area control means 14 calculates the duplication rate when the number of data storage areas 161 reserved in the prefetch buffer 16 exceeds a predetermined number (optional, for example, four). The means 13 is instructed to calculate the duplication rate of each generation in which the data storage area 161 is secured. Then, the data storage area control means 14 discards the data storage area 161 based on the calculation result. For example, the data storage area control means 14 discards the data storage area 161 for the generation having the lowest ratio of the data read in response to the read request among the data storage areas 161 reserved in the prefetch buffer 16. .. The data storage area control means 14 may be configured to exclude the most recently secured (newest) data reservation area 161 from the target to be discarded.

The data storage area control means 14 may determine the data storage area 161 to be discarded based on the duplication rate of each generation stored in the memory (not shown) calculated in advance by the duplication rate calculation means 13. Absent. Further, in the data storage area control means 14, for example, the ratio used to the capacity of the prefetch buffer 16 exceeds a preset standard (for example, a predetermined reference threshold value, which may be an arbitrary value). The data storage area 161 may be configured to be destroyed at a timing or the like.

In this way, the data storage area control means 14 newly secures or secures the data storage area 161 in the prefetch buffer 16 based on the instruction from the data reading means 12 and the duplication rate calculated by the duplication rate calculation means 13. The data storage area 161 that has been created is discarded.

As described above, the data reading means 12, the duplication rate calculating means 13, and the data storage area control means 14 cooperate with each other to newly secure or secure the data storage area 161 in the prefetch buffer 16. The storage area 161 is discarded, or the pre-read fragment data is stored in the reserved data storage area 161. In other words, the data read control means (data read control unit) including the data read means 12, the overlap rate calculation means 13, and the data storage area control means 14 is based on the overlap rate calculated by the duplicate rate calculation means 13. , It is also possible to control the storage status of fragment data for each generation stored in the prefetch buffer 16.

The disk device 15 is a storage device such as a hard disk. Fragment data is stored in the disk device 15 in a state of deduplication in units of data blocks. Although only one disk device 15 is shown in FIG. 3, as described above, the disk device 15 is actually composed of a storage device such as a hard disk provided by each of the storage nodes 3 and the like. There is.

The prefetch buffer 16 is a storage device such as a memory. A data storage area 161 can be secured in the prefetch buffer 16 for each generation. The fragment data read ahead by the data reading means 12 is stored in the data storage area 161 according to the generation to which the fragment data belongs. In this way, the prefetch buffer 16 temporarily stores the fragment data read from the disk device 15.

The above is an example of the configuration of the storage system 1. Subsequently, an example of the flow when the storage system 1 reads the fragment data from the disk device 15 will be described with reference to FIGS. 7 and 8. First, with reference to FIG. 7, an example of an overall flow when the storage system 1 reads fragment data from the disk device 15 will be described.

For example, the storage system 1 receives the input of the backup image file requested to be read. Then, the data reading means 12 starts the process of reading the data in order from the start of the backup image file.

For example, the data reading means 12 attempts to restore the data block from the beginning of the file. Since the data block is divided into a plurality of fragment data and stored, the data reading means 12 reads the fragment data in order. For example, the data reading means 12 acquires management information from the disk device 15. Then, the data reading means 12 refers to the acquired management information and specifies the storage file and the position of the fragment data generated based on the data block (step S101).

Subsequently, the data reading means 12 reads the fragment data. First, the data reading means 12 confirms whether or not the fragment data to be read has already been read ahead and stored in the prefetch buffer 16 (step S102).

When the fragment data to be read is stored in the prefetch buffer 16 (step S102, Yes), the data reading means 12 acquires the target fragment data from the prefetch buffer 16 (step S103). In other words, when the fragment data to be read is stored in the data storage area 161, the data reading means 12 does not read from the disk device 15.

On the other hand, when the fragment data to be read is not stored in the prefetch buffer 16 (step S102, No), in the data reading means 12, the generation to which the fragment data to be read belongs belongs to the previously read fragment data. It is confirmed whether or not it is the same as the generation (step S105). The data reading means 12 determines the generation of the fragment data based on, for example, the information indicating the storage file and the position of the fragment data included in the management information.

When the generation to which the fragment data to be read belongs is not the same as the generation to which the previously read fragment data belongs (step S105, No), that is, when reading a new generation fragment, the data reading means 12 has an overlap rate. The calculation means 13 is instructed to calculate the duplication rate of the generation before the change. Further, the data reading means 12 instructs the data storage area control means 14 to secure the data storage area 161 for the changed generation. As a result, the duplication rate calculating means 13 calculates the duplication rate of the generation before the change, and the data storage area control means 14 secures the data storage area for the generation after the change (step S106).

The data storage area control means 14 confirms whether or not the duplication rate of the generation before the change calculated by the duplication rate calculation means 13 is lower than the predetermined discard reference value (step S107). When the duplication rate of the generation before the change calculated by the duplication rate calculating means 13 is lower than the discard reference value (step S107, Yes), the data storage area control means 14 sets the data storage area 161 for the generation before the change. Discard (step S108). On the other hand, when the duplication rate of the generation before the change calculated by the duplication rate calculating means 13 is equal to or greater than the discard reference value (step S107, No), the data storage area control means 14 has the data storage area 161 for the generation before the change. Do not discard.

Further, the data storage area control means 14 confirms whether or not the number of data storage areas 161 reserved in the prefetch buffer 16 exceeds a predetermined number (step S109). When the number of data storage areas 161 exceeds a predetermined number (step S109, Yes), the data storage area control means 14 discards the data storage area 161 of the generation having the lowest duplication rate (step S110).

When the generation to which the fragment data to be read belongs is the same as the generation to which the previously read fragment data belongs (steps S105, Yes), or after the processing of step S110, the number of data storage areas is less than or equal to a predetermined number. In some cases (step S109, No), the data reading means 12 reads the fragment from the disk device 15 (step S111). The process of step S111 may be performed in parallel with the process of step S107, for example. Details of the process in step S111 will be described later.

After the processing of step S111 or after the processing of step S103, the data reading means 12 confirms whether or not all the fragment data to be read has been read. When all the fragment data to be read is read (step S104, Yes), the data reading means 12 ends the process. On the other hand, when the fragment data to be read still remains, the data reading means 12 proceeds to read the next fragment data.

The above is an example of the overall flow when the storage system 1 reads fragment data from the disk device 15. Subsequently, the process of step S111 will be described in more detail with reference to FIG.

Referring to FIG. 8, the data reading means 12 reads the fragment data to be read from the disk device 15 (step S201).

Further, the data reading means 12 confirms whether or not the duplication rate of the generation to which the fragment data read in step S201 belongs is lower than the look-ahead reference value (step S202). Then, when the duplication rate of the generation to which the fragment data read in step S201 belongs is lower than the look-ahead reference value (step S202, Yes), the data reading means 12 does not perform look-ahead. On the other hand, when the duplication rate of the generation to which the fragment data read in step S201 belongs is equal to or higher than the look-ahead reference value (step S202, No), the data reading means 12 continues the fragment data to be read on the disk device 15. A predetermined number of subsequent fragment data is pre-read, for example, in the background. Then, the data reading means 12 stores a predetermined number of pre-read subsequent fragment data in the data storage area 161 of the corresponding generation of the prefetch buffer 16 (step S203).

The above is a detailed example of step S111 in FIG.

As described above, the storage system 1 in the present embodiment has the duplication rate calculation means 13 and the data storage area control means 14. With such a configuration, when the generation of the fragment data to be read changes, the data storage area control means 14 is based on the duplication rate of the generation before the change calculated by the duplication rate calculation means 13, and the data for the generation before the change. It is possible to decide whether or not to discard the storage area 161. As a result, it becomes possible to continue to retain the look-ahead data of the generation that is determined to have a high possibility that a large amount of data is stored based on the read data. As a result, in the storage system 1 having a function of eliminating duplicate storage, efficient reading can be performed, and deterioration of reading performance can be suppressed.

Further, the storage system 1 has a data reading means 12 for determining whether or not to read ahead based on the duplication rate calculated by the duplication rate calculating means 13. As a result, it becomes possible to pre-read the data of a generation that is more necessary, and it becomes possible to perform an efficient pre-reading. As a result, in the storage system 1 having a function of eliminating duplicate storage, efficient reading can be performed, and deterioration of reading performance can be suppressed.

It should be noted that the predetermined look-ahead reference value used when the data reading means 12 determines whether or not to perform look-ahead and whether or not the data storage area control means 14 discards the data storage area 161 for the generation before the change. The discard reference value used in the determination may be the same value or a different value.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIG. In the second embodiment, the outline of the configuration of the storage device 6 will be described.

Referring to FIG. 9, the storage device 6 includes a data storage unit 61, a temporary data storage unit 62, and a data read control unit 63. For example, the storage device 6 has an arithmetic unit (not shown) and a storage device (not shown), and the arithmetic unit executes a program stored in the storage device to realize the data read control unit 63.

The data storage unit 61 is a storage device such as a hard disk. The data storage unit 61 stores data of a plurality of generations by deduplication.

The temporary data storage unit 62 is a storage device such as a memory. The temporary data storage unit 62 temporarily stores the data read from the data storage unit 61.

The data read control unit 63 reads the data stored in the data storage unit 61 and stores it in the temporary data storage unit 62. Further, the data read control unit 63 reads data from the temporary data storage unit 62. As described above, the data read control unit 63 is configured to read the data stored in the data storage unit 61 and store it in the temporary data storage unit 62, and also read the data from the temporary data storage unit 62.

Further, the data read control unit 63 controls the data stored in the temporary data storage unit 62 based on the ratio of the data for each generation to the data read in response to the read request received from the external device.

As described above, the storage device 6 in the present embodiment has the data read control unit 63. With such a configuration, the data read control unit 63 can control the data stored in the temporary data storage unit 62 based on the ratio of the data for each generation to the data read in response to the read request. As a result, the temporary data storage unit 62 can be efficiently used. As a result, in the storage system 1 having a function of eliminating duplicate storage, efficient reading can be performed, and deterioration of reading performance can be suppressed.

Further, the above-mentioned storage device 6 can be realized by incorporating a predetermined program into the storage device 6. Specifically, the program according to another embodiment of the present invention has a data storage unit 61 that deduplications and stores data of a plurality of generations, and temporary data that temporarily stores data read from the data storage unit 61. A data read control means (data read control) in which the storage device 6 having the storage unit 62 reads the data stored in the data storage unit 61 and stores the data in the temporary data storage unit 62, and also reads the data from the temporary data storage unit 62. Unit 63) is realized, and the data read control means controls the data stored in the temporary data storage unit 62 based on the ratio of the data for each generation to the data read in response to the read request received from the external device. It is a program.

Further, the information processing method executed by the storage device 6 described above includes a data storage unit 61 that deduplications and stores data of a plurality of generations, and a temporary storage unit 61 that temporarily stores the data read from the data storage unit 61. An information processing device (storage device 6) having the data storage unit 62 reads out the data stored in the data storage unit 61 and stores the data in the temporary data storage unit 62, and also reads the data from the temporary data storage unit 62 and externally. This is a method of controlling the data stored in the temporary data storage unit 62 based on the ratio of the data for each generation to the data read in response to the read request received from the device.

Even the invention of the program or the information processing method having the above-mentioned configuration can achieve the above-mentioned object of the present invention because it has the same operation as the above-mentioned storage device 6.

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

(Appendix 1)
A data storage unit that deduplications and stores data of multiple generations,
A temporary data storage unit that temporarily stores the data read from the data storage unit, and
A data read control unit that reads data stored in the data storage unit and stores the data in the temporary data storage unit, and also reads data from the temporary data storage unit.
Have,
The data read control unit controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. Storage device.
(Appendix 2)
The storage device according to Appendix 1.
The data read control unit is configured to secure a data storage area for each generation in the temporary data storage unit and store the data of the generation corresponding to the secured data storage area, and the data to be read in response to the read request. When the generation to which the change belongs, the data storage area for the generation after the change is secured, and the data before the change is based on the ratio of the data of the generation before the change to the data read in response to the read request. A storage device that determines whether or not to discard the data storage area for the generation of.
(Appendix 3)
The storage device according to Appendix 2,
The data read control unit is used for the generation before the change when the value obtained by dividing the size of the data of the generation before the change by the size of the data read in response to the read request is lower than a predetermined reference value. A storage device that discards the data storage area of the above.
(Appendix 4)
The storage device according to any one of Supplementary note 1 to 3.
When the number of the data storage areas reserved in the temporary data storage unit exceeds a predetermined number predetermined, the data read control unit of each generation of the data read in response to the read request. A storage device that discards the data storage area based on the percentage of data occupied.
(Appendix 5)
The storage device according to Appendix 4,
When the number of the data storage areas reserved in the temporary data storage unit exceeds a predetermined number, the data read control unit of the data read control unit secures the data storage area in the temporary data storage unit. A storage device that discards the data storage area of the generation having the lowest ratio contained in the read data.
(Appendix 6)
The storage device according to any one of Appendix 1 to 5.
When the data read control unit reads the data stored in the data storage unit based on the ratio of the data of each generation to the data read in response to the read request, the data read control unit reads the data to be read and also reads the data to be read. A storage device that determines whether or not to read a predetermined range of data following the data to be read.
(Appendix 7)
The storage device according to Appendix 6.
When the ratio of the generation to which the data to be read belongs to the data read in response to the read request is equal to or more than a predetermined reference value, the data read control unit reads the data stored in the data storage unit. A storage device that reads data to be read and a predetermined range of data following the data to be read.
(Appendix 8)
The storage device according to Appendix 6 or 7.
The data read control unit is a storage device that, when reading data in a predetermined range following the data to be read, stores the read data in the predetermined range in the temporary data storage unit.
(Appendix 9)
A storage device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
A data read control means for reading data stored in the data storage unit and storing the data in the temporary data storage unit and reading data from the temporary data storage unit is realized.
The data read control means controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. Program to do.
(Appendix 9-1)
The program described in Appendix 9
The data read control means is configured to secure a data storage area for each generation in the temporary data storage unit and store the data of the generation corresponding to the secured data storage area, and the data to be read in response to the read request. When the generation to which the change belongs, the data storage area for the generation after the change is secured, and the data before the change is based on the ratio of the data of the generation before the change to the data read in response to the read request. A program that determines whether or not to discard the data storage area for the generation of.
(Appendix 9-2)
The program described in Appendix 9-1.
The data read control means is used for the generation before the change when the value obtained by dividing the size of the data of the generation before the change by the size of the data read in response to the read request is lower than a predetermined reference value. A program that destroys the data storage area.
(Appendix 10)
An information processing device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
The data stored in the data storage unit is read and stored in the temporary data storage unit, the data is read from the temporary data storage unit, and the data of each generation among the data read in response to the read request received from the external device. An information processing method that controls the storage status of data for each generation stored in the temporary data storage unit based on the ratio of the data.
(Appendix 10-1)
The information processing method described in Appendix 10
A data storage area for each generation is secured in the temporary data storage unit, data of the generation corresponding to the secured data storage area is stored, and the data is changed when the generation to which the data to be read belongs changes in response to the read request. The data storage area for the later generation is secured, and the data storage area for the previous generation is discarded based on the ratio of the data of the previous generation to the data read in response to the read request. An information processing method that determines whether or not to do so.
(Appendix 10-2)
The information processing method according to Appendix 10-1.
When the value obtained by dividing the size of the data of the generation before the change by the size of the data read in response to the read request is lower than the predetermined reference value, the data storage area for the generation before the change is discarded. Information processing method.

The programs described in each of the above embodiments and appendices may be stored in a storage device or recorded in 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 present invention has been described above with reference to each of the above embodiments, 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.

1 Storage system 11 Data writing means 12 Data reading means 13 Duplicate rate calculation means 14 Data storage area control means 15 Disk device 16 Prefetch buffer 161 Data storage area 2 Accelerator node 3 Storage node 4 Backup system 5 Backup target device 6 Storage device 61 Data Storage unit 62 Temporary data storage unit 63 Data read control unit

Claims (10)

A data storage unit that deduplications and stores data of multiple generations,
A temporary data storage unit that temporarily stores the data read from the data storage unit, and
A data read control unit that reads data stored in the data storage unit and stores the data in the temporary data storage unit, and also reads data from the temporary data storage unit.
Have,
The data read control unit controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. Storage device. The storage device according to claim 1.
The data read control unit is configured to secure a data storage area for each generation in the temporary data storage unit and store the data of the generation corresponding to the secured data storage area, and the data to be read in response to the read request. When the generation to which the change belongs, the data storage area for the generation after the change is secured, and the change is based on the ratio of the data of the previous generation to the data read in response to the read request. A storage device that determines whether to discard the data storage area for the previous generation. The storage device according to claim 2.
The data read control unit is used for the generation before the change when the value obtained by dividing the size of the data of the generation before the change by the size of the data read in response to the read request is lower than a predetermined reference value. A storage device that discards the data storage area of the above. The storage device according to claim 2 or 3.
When the number of the data storage areas reserved in the temporary data storage unit exceeds a predetermined number predetermined, the data read control unit of each generation of the data read in response to the read request. A storage device that discards the data storage area based on the percentage of data occupied. The storage device according to claim 4.
When the number of the data storage areas reserved in the temporary data storage unit exceeds a predetermined number, the data read control unit of the data read control unit secures the data storage area in the temporary data storage unit. A storage device that discards the data storage area of the generation having the lowest ratio contained in the read data. The storage device according to any one of claims 1 to 5.
When the data read control unit reads the data stored in the data storage unit based on the ratio of the data of each generation to the data read in response to the read request, the data read control unit reads the data to be read and also reads the data to be read. A storage device that determines whether or not to read a predetermined range of data following the data to be read. The storage device according to claim 6.
When the ratio of the generation to which the data to be read belongs to the data read in response to the read request is equal to or more than a predetermined reference value, the data read control unit reads the data stored in the data storage unit. A storage device that reads data to be read and a predetermined range of data following the data to be read. The storage device according to claim 6 or 7.
The data read control unit is a storage device that, when reading data in a predetermined range following the data to be read, stores the read data in the predetermined range in the temporary data storage unit. A storage device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
A data read control means for reading data stored in the data storage unit and storing the data in the temporary data storage unit and reading data from the temporary data storage unit is realized.
The data read control means controls the storage status of the data for each generation stored in the temporary data storage unit based on the ratio of the data of each generation to the data read in response to the read request received from the external device. Program to do. An information processing device having a data storage unit that deduplications and stores data of a plurality of generations and a temporary data storage unit that temporarily stores data read from the data storage unit.
The data stored in the data storage unit is read and stored in the temporary data storage unit, the data is read from the temporary data storage unit, and the data of each generation among the data read in response to the read request received from the external device. An information processing method that controls the storage status of data for each generation stored in the temporary data storage unit based on the ratio of the data.

Images