JP6926866B2 - Storage controller and storage control program - Google Patents

Description

The present invention relates to a storage control device and a storage control program.

Conventionally, there is a garbage collection process that collects unused data areas and secures a continuous data area. As a related prior art, for example, image data is divided into a size equal to or smaller than the size of a plurality of storage areas having the same size and at equal intervals divided by a hard disk, and compressed. In addition, a technology that presents candidates for compression target files to the user based on file management information and user settings, compresses the specified file as the compression target file, and overwrites the compressed file with the original file. There is.

Japanese Unexamined Patent Publication No. 9-1253539 Japanese Unexamined Patent Publication No. 2001-43115

However, according to the prior art, if the compressed data is written in advance without dividing the storage area of the storage, the garbage collection process is performed, and the CPU (Central Processing) that performs the garbage collection process is performed. Unit) load and storage load increase. Then, due to the increase in load, the performance of the storage during the garbage collection process deteriorates. Further, when the compressed data is written in the data area in which the storage area of the storage is divided at equal intervals, the smaller the compressed data with respect to the above-mentioned data area, the larger the area in which the data is not written. Here, even if a plurality of data of the same size are compressed, the size of the compressed data differs depending on the content of the data. Therefore, the storage area of the storage is not sufficiently used.

In one aspect, it is an object of the present invention to provide a storage control device and a storage control program that can efficiently use the storage for writing the compressed data.

In one embodiment, the storage controller, and the storage control program, write data and write data of a predetermined size for storage having a storage area including a plurality of data areas that are integral multiples of the size of the access unit of the storage. Indicates information that can identify the physical address of each data area of a plurality of data areas of a first size that is an integral multiple of the data area allocated from the storage area and the usage status of each data area according to the reception of the logical address of. First management information having information, and information capable of identifying the physical address of each data area of a plurality of data areas having a second size different from the first size, which is an integral multiple allocated from the storage area. From the second management information having the information indicating the usage status of each data area, the management information corresponding to the size larger than the size of the compressed data obtained by compressing the write data is specified, and the specified management information is specified. The compressed data is written to the unused data area among the plurality of data areas to be created, and the storage area is associated with the information that can identify the physical address of the unused data area and the logical address of the written data. The physical address of the data area in which the data is written is stored in the logical address information in which the logical address of the data is associated with the identifiable information.

In one aspect, the invention allows efficient use of storage for writing compressed data.

FIG. 1 is an explanatory diagram showing an operation example of the storage control device 101 according to the first embodiment. FIG. 2 is an explanatory diagram showing a hardware configuration example of the storage system 200. FIG. 3 is an explanatory diagram showing an example of functional configuration of CM211. FIG. 4 is an explanatory diagram showing an example of the stored contents of the block map table 311. FIG. 5 is an explanatory diagram showing an example of the stored contents of the data size management table 312-1. FIG. 6 is an explanatory diagram (No. 1) showing an operation example at the time of receiving Write I / O. FIG. 7 is an explanatory diagram (No. 2) showing an operation example at the time of receiving Write I / O. FIG. 8 is an explanatory diagram (No. 3) showing an operation example at the time of receiving Write I / O. FIG. 9 is an explanatory diagram showing an operation example of the reference counter addition / subtraction process. FIG. 10 is an explanatory diagram showing an operation example when the Read I / O is accepted. FIG. 11 is a flowchart showing an example of the processing procedure at the time of receiving Write I / O. FIG. 12 is a flowchart showing an example of the reference counter addition / subtraction processing procedure. FIG. 13 is a flowchart showing an example of the Read I / O reception processing procedure. FIG. 14 is an explanatory diagram showing a state of the container when the data pattern tendency changes. FIG. 15 is an explanatory diagram showing an example of functional configuration of CM211 according to the second embodiment. FIG. 16 is an explanatory diagram showing an example of the stored contents of the data size management table 312-x. FIG. 17 is an explanatory diagram (No. 1) showing an example of physical release of the physical release unit area of the container area 121. FIG. 18 is an explanatory diagram (No. 2) showing an example of physical release of the physical release unit area of the container area 121. FIG. 19 is an explanatory diagram (No. 3) showing an example of physical release of the physical release unit area of the container area 121. FIG. 20 is a flowchart showing an example of the reference counter addition / subtraction processing procedure according to the second embodiment. FIG. 21 is a flowchart (No. 1) showing an example of the garbage collection processing procedure according to the second embodiment. FIG. 22 is a flowchart (No. 2) showing an example of the garbage collection processing procedure according to the second embodiment.

Hereinafter, embodiments of the disclosed storage control device and storage control program will be described in detail with reference to the drawings.

(Embodiment 1)
FIG. 1 is an explanatory diagram showing an operation example of the storage control device 101 according to the first embodiment. The storage control device 101 is a computer that controls the storage 102. Specifically, the storage control device 101 provides the storage area of the storage 102 to the user. The storage control device 101 is, for example, a server. The storage control device 101 can improve the usage efficiency of the storage 102 by dividing the data from the user, compressing each of the divided data, and writing the compressed data to the storage 102.

Here, if the storage area of the storage 102 is a storage area accessed by the head, for example, if the storage 102 is an HDD (Hard Disk Drive), it is preferable to pre-write each compressed data. The reason why it is preferable to write each compressed data in a pre-filled manner is that the HDD can be accessed at a higher speed as it is located on the outer periphery of the magnetic disk, and it is a sequential access because it accesses a continuous area. Further, the usage efficiency of the storage 102 can be improved by writing the compressed data without opening the space between them.

Here, if data is overwritten when the data is pre-packed, it is possible to treat the old data already written in the storage as unused and then write the new data to the storage. can. When this operation is performed, the unused area on the storage will increase. As the unused area increases, the storage usage efficiency decreases. Therefore, it is conceivable to perform garbage collection processing in order to eliminate the unused area. In the garbage collection process, data other than unused data is pre-packed and written again.

However, when the garbage collection process is performed, the data is rearranged, so that the load on the storage and the load on the CPU that performs the garbage collection process increase. In addition, since the memory is used as a temporary storage location for the data read from the storage for data relocation, the memory is temporarily consumed. Then, due to the increase in load, the performance of the storage during the garbage collection process deteriorates.

Further, in order to suppress the execution of the garbage collection process, it is conceivable to write the compressed data to the data area in which the storage area of the storage is divided at equal intervals. However, even if a plurality of data of the same size are compressed, the size of the compressed data differs depending on the content of the data. Therefore, when the compressed data is written to the data area in which the storage area of the storage is divided at equal intervals, the smaller the compressed data with respect to the above-mentioned data area, the larger the area where the data is not written. Storage storage area is not fully used.

Therefore, in the present embodiment, a data area larger than the size of the compressed data obtained by compressing the write data is specified from each management information that manages an area group that is an integral multiple of each access unit of the storage 102, and the data area is set to the above-mentioned data area. Writing the above-mentioned compressed data will be described.

An operation example of the storage control device 101 will be described with reference to FIG. The storage control device 101 controls the storage 102. Then, the storage control device 101 can access the management information 110 and the logical address information 111. The management information 110 and the logical address information 111 are, for example, in the memory of the storage control device 101, but may be in the storage area of another device.

The management information 110 has the following two pieces of information corresponding to the first size and the second size, which are integral multiples of the size of the access unit of the storage 102. The first information is information that can identify the physical address of each data area of the plurality of data areas of the first size or the second size allocated from the storage area of the storage 102. The second information is information indicating the usage status of each data area. The integer multiple of the size of the access unit can be any multiple. In the example of FIG. 1, it is assumed that the size of the access unit is 512 [byte]. Then, in the example of FIG. 1, the storage control device 101 corresponds to the management information 110-1 and the second size different from the first size as the first management information corresponding to the first size. As the management information of 2, the management information 110-2 can be accessed. In the example of FIG. 1, the first size is 3584 [byte], which is 7 times 512 [byte], and the second size is 1536 [byte], which is 3 times 512 [byte].

In the following description, reference codes such as "management information 110-1" and "management information 110-2" are used to distinguish the same type of elements, and "management information 110-2" is used when the same type of elements are not distinguished. Only the common number among the reference codes may be used as in "Management information 110".

Further, a plurality of data areas that are integral multiples of the size of the access unit allocated from the storage area of the storage 102 are referred to as "container areas". There are a plurality of data areas 122 in the container area 121. The number of data areas 122 may be any number as long as it is an integer of 2 or more, and is, for example, 1 [M pieces]. Further, the data area 122 is preferably a continuous area so that it can be accessed sequentially. In the example of FIG. 1, the container area 121-1 corresponding to the management information 110-1 has a plurality of data areas 122-1 to 2 of 3584 [byte]. Further, the container area 121-2 corresponding to the management information 110-2 has a plurality of data areas 122-20 to 2 of 1536 [byte]. Further, the management information 110 has a container ID (Identifier) that identifies the container area 121 in order to identify the corresponding container area 121. In the example of FIG. 1, it is assumed that the container ID of the container area 121-1 is 1 and the container ID of the container area 121-2 is 2.

Further, the information that can identify the physical address of each data area 122 included in the container area 121 may be the physical address itself of each data area 122, the address of the head data area of the plurality of data areas, and the information of the data area. It may be the size. In the example of FIG. 1, the management information 110 has a container start address 112 of the container area 121 and a container management size 113 as the size of the data area.

Further, the information indicating the usage status of the data area 122 is specifically one of the following three identifiers. The first identifier is "empty", which indicates a situation in which the corresponding data area 122 has never been used since it was secured. The second identifier is "in use" indicating a situation in which data is stored in the corresponding data area 122 and the data area 122 is being used. The third identifier is "unused", which indicates a situation in which data has been stored in the corresponding data area 122 in the past, but no data is currently stored. In the example of FIG. 1, the information indicating the usage status of the data area 122 is referred to as "data area usage status". Then, in the example of FIG. 1, the management information 110 has container information 114 having a data area usage status of each data area 122 and a slot number indicating the order from the first data area in each data area. The slot number may start from 0 or 1, but it is preferable to start from 0 in order to specify the physical address of each data area. The container information 114-1 shown in FIG. 1 has records 115-1-0 to 2. Further, the container information 114-2 has records 115-2-2 to 2.

The logical address information 111 is information in which the physical address of the data area 122 in which the data is written in the storage area of the storage 102 can be specified and the logical address of the above-mentioned data is associated with the information. The information that can identify the physical address of the data area 122 may be the physical address of the data area 122 itself, or may be a container number and a slot number. The physical address of the head data area 122 of the container area 121 to which the corresponding data area 122 belongs and the container management size 113 of the corresponding container area 121 can be specified by the container number. Next, the offset from the corresponding head data area 122 to the physical address of the data area 122 to be specified can be specified by the slot number and the container management size 113. Then, the physical address of the data area 122 to be specified can be specified by adding the above-mentioned offset to the physical address of the corresponding head data area 122.

The logical address information 111 shown in FIG. 1 has a logical address, a container ID, and a slot number. The logical address information 111 shown in FIG. 1 has records 116-0 and 1. Record 116-0 indicates that the data whose logical address is 0 [KB] is stored in the container ID = 1 and the slot number = 0. Further, record 116-1 indicates that the data whose logical address is 0 [KB] is stored in the container ID = 1 and the slot number 1.

First, the storage control device 101 receives a write request including write data of a predetermined size and a logical address of the write data. The predetermined size may be any size as long as it is a size equal to or larger than the access unit of the storage 102. For example, the predetermined size is 4096 [byte], 8192 [byte], or the like. In the following example, the predetermined size is assumed to be 4096 [byte]. Further, the write request is referred to as "Write I / O". Further, the written data is referred to as "Write data". Similarly, the read request is referred to as "Read I / O". The Read I / O includes the logical address of the read range. The read range is referred to as a "Read range".

Further, the storage control device 101 may accept the Write I / O including the Write data having a predetermined size or more. In this case, the storage control device 101 may divide the Write data included in the Write I / O into predetermined data, and may accept the divided predetermined Write data and the logical address of the divided predetermined Write data. .. Further, the issuer of Write I / O is a user of the storage control device 101, for example, a host device, but is not limited to this. For example, suppose that data is transferred from the storage to which the present embodiment is not applied to the storage 102 to which the present embodiment is applied. In this case, the device for migrating the data may issue the Write I / O to the storage control device 101 as the write data of the data read from the storage to which the present embodiment is not applied.

As shown in FIG. 1 (1), the storage control device 101 receives the Write data wd. The first logical address of the received Write data is 0 [KB]. In the example of FIG. 1, since the write data wd is 4096 [KB] or more, which is a predetermined size, the storage control device 101 divides the write data wd into 4096 [KB] units. Hereinafter, the data obtained by dividing the Write data will be referred to as "divided Write data". In the example of FIG. 1, as shown in FIG. 1 (2), the storage control device 101 divides the Write data wd to obtain the divided Write data wds-0, 1, ... Further, the logical address of the divided Write data wds-0 is 0 [KB], and the logical address of the divided Write data wds-1 is 4 [KB]. Here, since the logical addresses 0 [0KB] and 4 [KB] are the addresses registered in the logical address information 111, the data is overwritten.

Next, the storage control device 101 compresses the divided Write data wds in response to receiving the divided Write data wds having a predetermined size and the logical address of the divided Write data wds. Hereinafter, the compressed data obtained by compressing the divided Write data will be referred to as "compressed divided Write data". In the example of FIG. 1, as shown in (3-1) and (3-2) of FIG. 1, the storage control device 101 compresses the divided Write data wds-1 and 2, respectively, and compresses the divided Write data wdsc. Get -0, 1. Here, it is assumed that the respective sizes of the compression division Write data wdsc-0 and 1 are 3584 [byte] and 1536 [byte].

Then, the storage control device 101 specifies the management information 110 corresponding to the size equal to or larger than the size of the compression division Write data wdsc from the management information 110. By specifying the management information 110, the container area 121 to be stored in the compression division Write data wdsc is determined. Here, the size equal to or larger than the size of the compression division Write data wdsc is preferably the size equal to or larger than the size of the compression division Write data wdsc and is the smallest size. In the example of FIG. 1, as shown in (4-1) of FIG. 1, since the size of the compression division Write data wdsc-0 is 3584 [byte], the storage control device 101 has a container management size of 3584 [byte]. ], The management information 110-1 is specified. Similarly, as shown in (4-2) of FIG. 1, since the size of the compression division Write data wdsc-1 is 1536 [byte], the storage control device 101 has a container management size of 1536 [byte]. Identify management information 110-2.

Here, at the time when the compression division Write data wdsc is obtained, there is a possibility that the management information 110 that is larger than the size of the compression division Write data wdsc and corresponds to the smallest size does not exist. In this case, the storage control device 101 may generate the management information 110 corresponding to the smallest size described above. More specifically, the storage control device 101 secures the container area 121 having the plurality of data areas 122 having the smallest size described above from the storage 102. Then, the storage control device 101 generates the management information 110 on the memory, and sets the container head address 112 of the generated management information 110 to the head addresses of the plurality of data areas 122 of the secured container area 121. Further, the storage control device 101 sets the container management size 113 of the generated management information 110 to the smallest size described above, and sets the data area usage status of all the records 115 of the container information 114 to "empty".

Further, even if the size of the compression division Write data wdsc is larger than the size of the compression division Witzc and the management information 110 corresponding to the smallest size exists, the data area usage status of all the records 115 of the container information 114 may be "in use". There is also sex. In this case as well, the storage control device 101 may generate new management information 110 corresponding to the smallest size described above. As described above, there may be a plurality of management information 110 corresponding to the same size.

Further, the management information 110 will be generated in n types from one times the size of the access unit to a predetermined size at the maximum. In the example of FIG. 1, since the size of the access unit is 512 [bytes] and the predetermined size is 4096 [bytes], n = 8.

Next, the storage control device 101 writes the compression division Write data wdsc to the unused data area 122 of the plurality of data areas 122 specified from the specified management information 110. Further, the storage control device 101 writes the compression division Write data wdsc and stores the information capable of specifying the physical address of the unused data area and the logical address of the division Write data in the logical address information 111 in association with each other. do.

Here, the data is overwritten, and the specified container ID and the container ID of the old data before overwriting indicated by the record 116 may be the same. In this case, the storage control device 101 may overwrite the data area 122 indicated by the slot number of the old data with the compression division Write data wdsc. As a result, the storage control device 101 does not have to update the data area usage status and the logical address information 111 of the container information 114. In the case of FIG. 1, the data is overwritten, and for the logical address 0 [KB], the specified container ID and the container ID of the old data before overwriting indicated by the record 116-0 are both 1, which are the same. be. In this case, as shown in (5-1) of FIG. 1, the storage control device 101 writes the compression division Write data wdsc in the data area 122-1-0 indicated by the slot number 0 of the old data.

Further, the logical address 4 [KB] is a different container ID because the specified container ID is 2 and the container ID of the old data before overwriting indicated by the record 116-1 is 2. In this case, as shown in (5-2) of FIG. 1, the storage control device 101 writes the compression division Write data wdsc to the unused data area 122-2-2 of the specified container ID = 2. Further, as shown in FIG. 1 (6-2), the storage control device 101 stores the container ID = 2 and the slot number 2 for the record 116-1. Further, the storage control device 101 updates the data area usage status of the record 115-1-1 with the container ID = 1 of the old data to "unused", and the data area of the record 115-2-2 with the container ID = 2. Update the usage status to "in use".

By the above processing, the storage control device 101 can efficiently use the storage 102. Specifically, the data areas 122 included in one container area 121 are all the same size. Therefore, in the garbage collection process, for example, the data in the rear data area 122 may be moved to the unused data area 122 in the front, and the other data areas 122 are not affected. Therefore, the storage control device 101 can suppress the load on the CPU that performs the garbage collection process and the load on the storage 102 as compared with the case where the data area has a variable length.

Further, the storage control device 101 uses the data area 122 suitable for the size of the compression division Write data wdsc, so that the storage area of the data area 122 is sufficiently used. Specifically, since the size of the data area 122 exists for every integral multiple of the access unit of the storage 102, the size of the area in which the compression division Write data wdsc is written is the size of the area where the data is not written. , Always less than 512 [byte]. On the other hand, when the compressed data is written to the data area in which the storage area of the storage is divided at equal intervals, the size of the area in which the compressed data is not written in one data area is less than the data size at equal intervals. .. Even if a plurality of data of the same size are compressed, the size of the compressed data differs depending on the content of the data. Therefore, in order to guarantee the case where the data size is hardly reduced even after compression, the data size at equal intervals is, for example, the data size before compression. Using the example of FIG. 1, since the data size before compression is 4096 [byte], the size of the area in which data is not written in one data area is less than 4096 [byte], and the present embodiment. Compared to, the storage area of the storage is not fully used.

Further, in FIG. 1, only two management information 110s, that is, management information 110-1 and 110-2, are displayed, but the storage control device 101 may be able to access three or more management information 110s. Then, the storage control device 101 that can access the three or more management information 110 identifies the management information 110 corresponding to the size equal to or larger than the size of the compression division Write data wdsc from the three or more management information 110.

Further, the method described with reference to FIG. 1 can be applied even if the storage 102 is a storage to which deduplication is applied. Here, deduplication detects the same data from a plurality of data in the storage, leaves one data among the plurality of the same data, excludes the other data, and the physical address of the remaining data. Corresponds to, manages the number of logical addresses that refer to the remaining data. An example in which the storage control device 101 is applied to the storage system will be described with reference to FIG.

FIG. 2 is an explanatory diagram showing a hardware configuration example of the storage system 200. The storage system 200 includes a disk array device 201 and a host device 202. The disk array device 201 includes one or more CMs (Control Modules) 211, one or more disks 212, and one or more CAs (Channel Adapter) 213. The CM 211 has a CPU 221, a DRAM 222, and a flash ROM (Read Only Memory) 223. Here, CM211 corresponds to the storage control device 101 shown in FIG. Further, the disk 212 corresponds to the storage 102 shown in FIG.

The disk array device 201 is a device that provides a user with a large-capacity volume in which a plurality of disks are bundled. The host device 202 is a host computer used by the user. Further, the host device 202 may be a Web server or the like that uses the disk array device 201.

The CM211 is a controller control unit that manages resources of the disk 212, CA213, CPU221, DRAM222, and the like. The disk 212 is a non-volatile storage device that stores user data and control information. For example, the disk 212 is an HDD or an SSD (Solid State Drive). All of the disks 212 may be HDDs, all may be SSDs, or HDDs and SSDs may be mixed.

The CPU 221 is an arithmetic processing unit that controls the entire CM 211. Further, the CPU 221 may have a plurality of cores. The DRAM 222 is a volatile memory used as a primary cache memory in the CM211. Therefore, the DRAM 222 stores a part of user data and control information. The flash ROM 223 is a non-volatile memory that stores the storage control program according to the present embodiment. As the storage medium of the non-volatile memory, for example, a NOR flash memory or a NAND flash memory can be adopted.

(Example of functional configuration of CM211)
FIG. 3 is an explanatory diagram showing an example of functional configuration of CM211. CM211 has a control unit 300. The control unit 300 includes a reception unit 301, a container area determination unit 302, a writing unit 303, a specific unit 304, and a reading unit 305. The control unit 300 realizes the functions of each unit by executing the program stored in the storage device by the CPU 221. Specifically, the storage device is, for example, the flash ROM 223 shown in FIG. Further, the processing results of each part are stored in the DRAM 222, the register of the CPU 221, the cache memory of the CPU 221 and the like.

Further, the CM 211 can access the storage unit 310. The storage unit 310 is, for example, a DRAM 222. The storage unit 310 has a block map table 311 and data size-specific management tables 312-1, 2, .... The block map table 311 corresponds to the logical address information 111 shown in FIG. The data size management table 312 corresponds to the management information 110 shown in FIG. Here, the data size management table 312 includes the number of logical addresses that refer to each data area. The stored contents of the block map table 311 are shown in FIG. The stored contents of the data size management table 312 are shown in FIG.

Further, the CM 211 can access the disk 212. Then, a container area 121 corresponding to the data size management table 312 is secured on the disk 212. Here, in FIG. 3, a plurality of container areas 121 are secured on one disk 212, but the present invention is not limited to this. For example, one container area 121 may span a plurality of discs 212. Further, each data area 122 of the plurality of data areas 122 included in the management table 312 for each data size has data that does not overlap with each other due to deduplication.

The reception unit 301 receives Write I / O and Read I / O.

The container area determination unit 302 divides the data wd if the Write data wd is 4096 [byte] or more in response to the reception unit 301 receiving the Write I / O, and obtains the divided Write data wds having a predetermined size. Then, the container area determination unit 302 specifies the data size-specific management table 312 corresponding to a size equal to or larger than the size of the compressed divided Write data wdsc obtained by compressing the divided Write data wds. By specifying the management table 312 for each data size, the container area 121 that is the storage destination of the compression division Write data wdsc is determined.

The writing unit 303 writes the compression division Write data wdsc to the unused data area 122 of the container area 121 determined by the container area determining unit 302. Further, the writing unit 303 stores the container ID of the data area 122 and the slot head in association with each other in the block map table 311 together with the writing of the compression division Write data wdsc described above. Further, if there are a plurality of unused data areas 122 in the container area 121 determined by the container area determination unit 302, the writing unit 303 may write the compression division Write data wdsc to the unused data area 122 at the head. good.

Further, in FIG. 1, when the specified container ID and the container ID of the old data before overwriting are the same, the compression division Write data wdsc is overwritten on the data area 122 indicated by the slot number of the old data. I explained that it is also good. When this method is applied to deduplication, the writing unit 303 further has 1 logical address that refers to the data area 122 indicated by the slot number corresponding to the logical address of the divided Write data wds in the block map table 311. Determine if it exists. If the number of the corresponding logical addresses is 1, the data area 122 indicated by the slot number of the old data is not referenced by a plurality of logical addresses and can be overwritten. Therefore, when the number of logical addresses that refer to the corresponding data area 122 is 1, the writing unit 303 overwrites the data area 122 indicated by the slot number of the old data with the compression division Write data wdsc.

The specific unit 304 divides if the logical address of the Read range is 4096 [byte] or more according to the reception of the Read I / O, and obtains a divided Read range having a predetermined size. Then, the specifying unit 304 refers to the block map table 311 to specify the container ID corresponding to the logical address of the divided Read range and the slot number.

The reading unit 305 calculates the physical address of the data area 122 in which the data corresponding to the logical address of the divided Read range is written, and reads the data from the calculated physical address of the disk 212. Specifically, the reading unit 305 describes the above-mentioned physical address based on the first physical address of the plurality of data areas 122 in the specified container ID, the specified slot number, and the container management size of the specified container ID. Is calculated. For example, the reading unit 305 calculates the above-mentioned physical address by the following equation (1).

Physical address of data area 122 corresponding to the divided Read range = Container start address of the specified container ID + (Container management size of the specified container ID x Specified slot number) ... (1)

It is assumed that the slot number in the equation (1) starts from 0. When the slot number starts from 1, the reading unit 305 may calculate by subtracting the slot number in the equation (1) by 1.

FIG. 4 is an explanatory diagram showing an example of the stored contents of the block map table 311. The block map table 311 shown in FIG. 4 has records 401-0 to 12.

The block map table 311 includes fields of a logical address, a physical storage container ID, and a physical storage slot number. A value indicating a logical address is stored in the logical address field. In the physical storage container ID field, the ID of the container in which the data specified by the corresponding logical address is stored is stored. The physical storage slot number field stores the slot number in which the data specified by the corresponding logical address is stored.

For example, record 401-0 indicates that the data specified by the logical address 0 [KB] is stored in the data area 122 managed by the slot number 0 of the container ID 2.

Next, as an example of the data size-based management table 312, an example of the stored contents of the data size-based management table 312-1 will be described with reference to FIG.

FIG. 5 is an explanatory diagram showing an example of the stored contents of the data size management table 312-1. The data size management table 312 has a container ID, container information 501, an unused queue 502, a container management size 503, and a container start address 504.

The container ID is the ID of the container managed by CM211. The container information 501 has information on the container managed by CM211. The container information 501 shown in FIG. 5 has records 511-0 to 7.

Specifically, the container information 501 includes fields such as a slot number, a data area usage status, and a reference counter. The slot number field stores the slot number in the corresponding container. In the data area usage status field, an identifier indicating the usage status of the data area 122 managed by the slot number is stored. The identifiers indicating the usage status are "empty", "in use", and "unused" shown in FIG.

The reference counter of the corresponding data area 122 is stored in the reference counter field. Specifically, the reference counter is the number of logical addresses that refer to the corresponding data area 122.

The unused queue 502 is a queue that manages a data area 122 whose usage status is "unused" in the corresponding container. The container management size 503 is the size of the data area 122 in the corresponding container area 121. The container head address 504 is the head physical address of the data area 122, which is the headmost in the container area 121.

Next, an operation example of CM211 at the time of receiving Write will be described with reference to FIGS. 6 to 8.

FIG. 6 is an explanatory diagram (No. 1) showing an operation example at the time of receiving Write I / O. As shown in FIG. 6 (1), the CM 211 divides the Write data wd included in the Write request received from the host device 202 into 4096 [byte] units. Subsequent processing is performed in units of divided Write data wds by dividing Write data wd. In the example of FIG. 6, CM211 divides the Write data wd to obtain the divided Write data wds-0 to 5, ....

Next, as shown in (2) of FIG. 6, CM211 refers to the block map table 311 and specifies the storage position of the old data for each divided Write data wds. A specific example of the storage position of the old data will be specifically described. For example, suppose that the Write request is a request to write the Write data wd to the first logical address 0 [KB]. In this case, the logical address of the divided Write data wds-0 is 0 [KB]. Then, the CM 211 specifies that the old data of the divided Write data wds-0 is the container ID 2 and the slot number 0 with reference to the record 401-0 shown in FIG. 4 for the divided Write data wds-0. Further, the logical address of the divided Write data wds-1 is 4 [KB]. In this case, the logical address of the divided Write data wds-1 is 0 [KB]. Then, the CM 211 specifies that the old data of the divided Write data wds-1 is the container ID 6 and the slot number 0 with reference to the record 401-1 shown in FIG. 4 for the divided Write data wds-1.

Then, as shown in FIG. 6 (3), the CM 211 determines whether or not there is data that overlaps with each divided Write data wds among the compressed data stored in the disk 212. Specifically, the CM 211 calculates the hash value of the original data of the compressed data, and when the hash value is stored in the disk 212, the calculated hash value is associated with the physical storage container ID and the physical storage slot number, and the DRAM 222. And so on. Then, the CM 211 calculates the hash value of each divided Write data wds and compares it with the stored hash value to determine whether or not there is duplicate data. Specifically, when the hash value of each divided Write data wds and the stored hash value match, the CM211 determines that there is duplicate data. On the other hand, when the hash value of each divided Write data wds and all the stored hash values are different, CM211 determines that there is no duplicated data. In the example of FIG. 6, CM211 determines that there is no duplicated data for the divided Write data wds-0 and 2, and determines that there is duplicated data for the divided Write data wds-1.

When it is determined that there is no duplicate data for each divided Write data wds, the CM 211 compresses each divided Write data wds and obtains each compressed divided Wite data wdsc as shown in (4-1) of FIG. .. In the example of FIG. 6, CM211 compresses the divided Write data wds-0 and 2, respectively, and obtains the compressed divided Write data wdsc-0 and 2. Here, the compression division Write data wdsc-0 shown in FIG. 6 becomes 3584 [byte], and the compression division Write data wdsc-2 becomes 1536 [byte].

On the other hand, when it is determined that there is duplicated data for each divided Write data wds, the reference counter for the duplicated data is added by 1 as shown in (4-2) of FIG. The addition and subtraction of the reference counter will be described with reference to FIG.

FIG. 7 is an explanatory diagram (No. 2) showing an operation example at the time of receiving Write I / O. The state shown in FIG. 7 is a state in which the process of (4-1) in FIG. 6 is completed. As shown in (1) of FIG. 7, the CM211 determines the disk write size according to the compressed data size of the compressed divided Write data wdsc, and compresses the container corresponding to the determined disk write size after compression. Determine as the data storage destination. Specifically, the CM211 determines the size of the compressed data size rounded up in units of 512 [bytes], which is an access unit, as the disk write size. Here, since the data before compression is 4096 [byte], the disk write size is 512xn [byte], where n is an integer of any of 1 to 8.

In the example of FIG. 7, the CM211 determines the storage destination of the divided Write data wds-0 to a container having a container ID = 1 having a container management size of 3584 [byte]. Similarly, the CM211 determines the storage destination of the divided Write data wds-2 to a container having a container ID = 2 having a container management size of 1536 [byte].

FIG. 8 is an explanatory diagram (No. 3) showing an operation example at the time of receiving Write I / O. The state shown in FIG. 8 shows the state after the process shown in (1) of FIG. 7 is completed. As shown in FIG. 8 (1), the CM 211 determines whether or not there is a data area 122 connected to the unused queue 502 of the container determined as the storage destination. In the example of FIG. 8, the CM 211 refers to the unused queue 502 of the container ID = 1 determined as the storage destination of the divided Write data wds-0, and has a data area 122 connected to the unused queue 502. to decide. Further, the CM 211 refers to the unused queue 502 having the container ID = 2 determined as the storage destination of the divided Write data wds-2, and determines that there is no data area 122 connected to the unused queue 502.

Next, when the CM211 determines that there is a data area 122 connected to the unused queue 502, as shown in FIG. 8 (2-1), the CM211 of the unused queue 502 of the container determined as the storage destination. The compression division Write data wdsc is written in the first data area 122. In the example of FIG. 8, the CM211 writes the compression division Write data wdsc-0 in the data area 122 managed by the slot number 1 which is the head of the unused queue 502 of the data size management table 312-1.

Further, it is assumed that the CM 211 determines that there is no data area 122 connected to the unused queue 502. In this case, as shown in (2-2) of FIG. 8, the CM211 acquires the empty data area 122 of the container determined as the storage destination, and compresses and divides the Write data wdsc in the acquired empty data area 122. Write. In the example of FIG. 8, the CM211 has the compression division Write data wdsc-2 in the data area 122 managed by the slot number 3 whose usage status is “empty” in the container information 501 of the data size management table 312-2. To write. Here, in FIG. 8, the area in which the compression division Write data wdsc-0 and 2 are written is shaded.

FIG. 9 is an explanatory diagram showing an operation example of the reference counter addition / subtraction process. The state shown in FIG. 9 is a state in which the respective processes of (2-1) and (2-2) of FIG. 8 have been completed. As shown in (1) of FIG. 9, CM211 adds the reference counter of the data area 122 in which the compression division Write data wdsc is written this time. Further, the CM 211 subtracts the reference counter of the data area 122 that was used before writing the compression division Write data wdsc this time.

In the example of FIG. 9, it is assumed that the data area 122 used before writing the compression division Write data wdsc-0 is the data area 122 managed by the slot number 5 of the data size management table 312-1. Further, it is assumed that the data area 122 used before writing the compression division Write data wdsc-2 is the data area 122 managed by the slot number 2 of the data size-specific management table 312-2. Therefore, the CM 211 increments the reference counter of the record 511-1 of the data size management table 312-1 and the record 511-3 of the data size management table 312-2 by 1. Further, the CM211 decrements the record 511-5 of the data size-based management table 312-1 and the reference counter of the record 511-2 of the data size-based management table 312-2 by one.

FIG. 10 is an explanatory diagram showing an operation example when the Read I / O is accepted. The state shown in FIG. 10 is a state in which a Read request is received from the host device 202 after the process of FIG. 9 (1) is completed. As shown in FIG. 10 (1), the CM 211 divides the Read range ra included in the Read request received from the host device 202 into 4096 [bytes]. Subsequent processing is performed in units of the divided Read range ras that divides the Read range ra. In the example of FIG. 10, CM211 divides the Read range ra to obtain the divided Lead range ras-0 to 3, .... Further, FIG. 10 shows the divided Read range ras-3.

Next, as shown in FIG. 10 (2), the CM 211 refers to the block map table 311 and refers to the container ID of the data area 122 in which the compressed division data of each division Read range ras is stored and the slot number. And identify. In the example of FIG. 10, in the CM211 with reference to the record 401-3 shown in FIG. 4, the container ID of the data area 122 in which the data of the divided Read range ras-3 is stored is 1, and the slot number is 5. Identify as.

Then, as shown in (3) of FIG. 10, the CM211 refers to the data size-specific management table 312 corresponding to the specified container ID, and refers to the data area in which the compressed division data of each division Read range ras is stored. Calculate the physical address of 122. Specifically, CM211 calculates the physical address of the data area 122 corresponding to each divided Read range ras according to the equation (1) shown in FIG.

In the example of FIG. 10, the CM211 calculates the physical address of the data area 122 corresponding to the divided Read range ras-3 as follows according to the equation (1).

Physical address of data area 122 corresponding to split Read range ras-3 = 0x2000000000 + (3584x5) = 0x20004600

Next, as shown in FIG. 10 (4), the CM 211 reads the compression division data from the calculated physical address on the disk 212. In the example of FIG. 10, CM211 reads the compression division data dsc-3 from the physical address 0x20004600. After that, the CM 211 decompresses the read compression division data dsc-3 and transmits the decompressed data to the host device 202.

Next, a flowchart showing the processing executed by the CM 211 will be described with reference to FIGS. 11 to 13.

FIG. 11 is a flowchart showing an example of the processing procedure at the time of receiving Write I / O. The CM211 receives the Write I / O from the host device 202 (step S1101). Next, the CM211 divides the Write data wd into 4096 [bytes] (step S1102). When the Write data wd is 4096 [byte] or more, the CM211 performs the processing of steps S1103 to S1105 for each of the divided Write data wds obtained by dividing the Write data wd. In the description of FIGS. 11 and 12, one of the divided Write data wds will be described.

Then, CM211 refers to the block map table 311 and specifies the storage position of the old data for the divided Write data wds (step S1103). Next, CM211 determines whether or not there is data that overlaps with the divided Write data wds (step S1104).

When there is no data that overlaps with the divided Write data wds (step S1104: no duplicate data), the CM 211 compresses the divided Write data wds (step S1105). Next, CM211 determines the disk write size according to the size of the compressed divided Write data wdsc obtained by compressing the divided Write data wds (step S1106). Then, the CM211 determines the container corresponding to the disk write size as the storage destination of the compression division Write data wdsc (step S1107). Next, the CM 211 determines whether or not there is a data area 122 connected to the unused queue of the container determined as the storage destination (step S1108).

If there is no data area 122 connected to the unused queue of the container determined as the storage destination (step S1108: no data area is registered in the unused queue), the CM 211 is an empty data area of the container determined as the storage destination. Obtain 122 (step S1109). Then, the CM 211 writes the compression division Write data wdsc in the acquired empty data area 122 (step S1110).

If there is a data area 122 connected to the unused queue of the container determined as the storage destination (step S1108: the data area is registered in the unused queue), the CM 211 is unused of the container determined as the storage destination. The compression division Write data wdsc is written in the data area 122 at the head of the queue (step S1111). Then, the CM 211 deletes the data area 122 in which the compression division Write data wdsc is written from the unused queue (step S1112).

After the processing of step S1110 or step S1112 is completed, or when there is data that overlaps with the divided Write data wds (step S1104: there is overlapping data), CM211 executes the reference counter addition / subtraction process (step S1113). The reference counter addition / subtraction process will be described with reference to FIG.

Then, CM211 updates the block map table 311 (step S1114). The process of step S1114 will be described more specifically. Here, the points to be updated are the physical storage container ID and the physical storage slot number of the record corresponding to the logical address of the divided Write data wds in the block map table 311. Hereinafter, the corresponding physical storage container ID and physical storage slot number will be referred to as "corresponding physical storage container ID and physical storage slot number to be updated".

For example, it is assumed that CM211 determines that there is duplicate data in the process of step S1104. In this case, the CM 211 updates the physical storage container ID and the physical storage slot number to be updated to the physical storage container ID and the physical storage slot number in which the duplicated data is stored.

Further, it is assumed that the CM 211 determines that the data area 122 is not registered in the unused queue for the process of step S1108. In this case, the CM 211 updates the physical storage container ID and the physical storage slot number to be updated with the physical storage container ID and the physical storage slot number of the acquired data area 122. On the other hand, it is assumed that the CM 211 determines that the data area 122 is registered in the unused queue for the process of step S1108. In this case, the CM 211 updates the physical storage container ID and the physical storage slot number to be updated to the physical storage container ID and the physical storage slot number of the data area 122 in which the compression division write data is written.

After the processing in step S1114 is completed, the CM211 ends the Write I / O acceptance processing.

FIG. 12 is a flowchart showing an example of the reference counter addition / subtraction processing procedure. The CM211 adds 1 to the reference counter of the data area 122 in which the current compression division Write data wdsc is written (step S1201). Further, the CM 211 subtracts 1 from the reference counter of the data area 122 that was used before writing the compression division write data this time (step S1202). Then, the CM211 determines whether or not there is a data area 122 in which the reference counter has become 0 by updating the reference counter this time (step S1203).

When there is a data area 122 in which the reference counter has become 0 (step S1203: there is a data area in which the reference counter has become 0), the CM 211 assumes that the target data area 122 is unused and puts the target data area in the unused queue. 122 is connected (step S1204). After the processing in step S1204 is completed, or when there is no data area 122 in which the reference counter has become 0 (step S1203: there is no data area in which the reference counter has become 0), the CM211 ends the reference counter addition / subtraction processing.

FIG. 13 is a flowchart showing an example of the Read I / O reception processing procedure. CM211 accepts Read I / O (step S1301). Next, CM211 divides the Read range ra into 4096 [bytes] (step S1302). When the Read range ra is 4096 [byte] or more, the CM211 performs the processes of steps S1303 to S1306 for each divided Read range ras obtained by dividing the Read range ra. In the description of FIGS. 11 and 12, one divided Read range ras of each divided Write data wds will be described.

Then, CM211 refers to the block map table 311 to specify the container ID and the slot number for the divided Read range ras (step S1303). Next, CM211 refers to the data size-specific management table 312 corresponding to the specified container ID, and calculates the physical address of the data area 122 corresponding to the divided Read range ras (step S1304). Then, the CM 211 reads the compression division data from the calculated address on the disk 212 (step S1305). Next, CM211 decompresses the read compression division data (step S1306). After the processing in step S1306 is completed, the CM 211 combines the divided data obtained by decompressing each compressed division data, transmits the combined data as Read data to the host device 202, and performs the Read I / O acceptance processing. finish.

As described above, the CM 211 according to the first embodiment specifies a data area 122 larger than the size of the compression division Write data wdsc from the data size management table 312, and the above-mentioned compression division Write is applied to the specified data area 122. Write the data wdsc. As a result, the CM 211 can efficiently use the disc 212. Specifically, as described in FIG. 1, the CM 211 can suppress the load on the CPU 221 and the load on the disk 212 as compared with the case where the data area 122 has a variable length. Further, the CM 211 can sufficiently use the storage area of the disk 212.

Further, when the disk 212 is an HDD, if there are a plurality of unused data areas 122 among the container areas 121 determined by the container area determination unit 302, the CM 211 is compressed into the unused data area 122 at the head. The divided Write data wdsc may be written. As a result, the CM211 does not require garbage collection processing, and can always continue Write I / O and Read I / O in the pre-filled state. Further, the above-mentioned effect may be obtained as long as the storage area of the disk 212 is accessed by the head, and for example, the disk 212 may be an optical disk or the like. If the disk 212 is an HDD, the head is a magnetic head. If the disk 212 is an optical disc, the head is an optical head.

Further, the CM211 may calculate the physical address of the data area corresponding to the divided Read range ras based on the equation (1). As a result, the CM 211 can easily identify the physical address to be read, so that the reading speed can be improved. Specifically, when the compressed data is written in front without dividing the storage storage area, the compressed data is variable, so it is used as the identifier of the data area that stores the compressed data. , A table associated with the physical address of the data area will be prepared. However, in the present embodiment, since the physical address to be read is obtained by the container management size × slot number, it is not necessary to refer to the above-mentioned table. Therefore, the CM211 can improve the reading speed.

Further, when deduplication is not performed, if the specified container ID and the container ID of the old data are the same, the CM211 overwrites the data area 122 indicated by the slot number of the old data with the compression division Write data wdsc. You may. As a result, the CM211 does not have to update the block map table 311 and the data size management table 312, so that the load on the CM211 can be suppressed.

Further, when deduplication is performed, the CM211 determines whether or not the reference counter of the slot number corresponding to the logical address of the divided Write data wds in the block map table 311 is 1, in addition to the condition when deduplication is not performed. to decide. Then, if the reference counter is 1, the CM 211 may overwrite the data area 122 indicated by the slot number of the old data with the compression division Write data wdsc. As a result, the CM211 does not have to update the block map table 311 and the data size management table 312, so that the load on the CM211 can be suppressed even with deduplication.

(Embodiment 2)
Next, the storage control device 101 (CM211) according to the second embodiment will be described. The same parts as those described in the first embodiment are designated by the same reference numerals, and the illustration and description thereof will be omitted.

Here, the tendency of the data size after compression may change due to the change in the characteristics (I / O pattern) of the data written from the host device 202, that is, the change in the data pattern tendency. In the first embodiment, under such an environment, the physical storage area of the disk 212 may be unnecessarily reserved for the container (container area 121).

FIG. 14 is an explanatory diagram showing a state of the container when the data pattern tendency changes. In FIG. 14, first, it is assumed that the data pattern tendency is an I / O pattern in which the write data is compressed to a data size of 6 blocks. In this case, the usage status of each slot (data area 122) in the 6-block container is as shown in (14-1) of FIG.

After that, it is assumed that the data pattern tendency changes to an I / O pattern in which the write data is compressed to a data size of 7 blocks. In this case, the usage status of each slot in the 6-block container and the usage status of each slot in the 7-block container are as shown in FIG. 14 (14-2).

In the state of the 6-block container as shown in (14-2), although most of the slots are unused, the physical capacity is consumed. As a result, both the 6-block container and the 7-block container consume physical capacity, and even if deduplication and data compression are performed, the effect of efficiently using the disk 212 may be diminished. be.

Therefore, in the second embodiment, the storage control device 101 (CM211) that determines whether or not garbage collection needs to be performed for each container and physically releases the unused area in the container to improve the usage efficiency of the disk 212. explain. First, a functional configuration example of the storage control device 101 (CM211) according to the second embodiment will be described.

(Example of functional configuration of CM211)
FIG. 15 is an explanatory diagram showing an example of functional configuration of CM211 according to the second embodiment. CM211 has a control unit 300. The control unit 300 includes a reception unit 301, a container area determination unit 302, a writing unit 303, a specific unit 304, a reading unit 305, and a physical release control unit 1501. The control unit 300 realizes the functions of each unit by executing the program stored in the storage device by the CPU 221. Further, the processing results of each part are stored in the DRAM 222, the register of the CPU 221, the cache memory of the CPU 221 and the like.

The physical release control unit 1501 controls the physical release of the physical release unit area of the container area 121. Here, the physical release unit area is an area in which a plurality of data areas 122 included in the container area 121 are divided into units that can be physically released. The physical release unit region is, for example, a region of 21 [MB].

Specifically, the physical release control unit 1501 determines whether or not the number of unused data areas 122 in the container area 121 exceeds the threshold value α. In the following description, the number of unused data areas 122 in the container area 121 may be referred to as “number of unused areas”. Here, the threshold value α can be arbitrarily set, and is set to, for example, a value of about 30% of the total number of data areas 122 included in the container area 121.

As an example, assuming that the storage capacity of the container area 121 is 4 [GB] and the storage capacity of the data area 122 is 512 [byte], the total number of data areas 122 of the container area 121 is "838608". In this case, the threshold value α is set to, for example, a value “2768240” which is about 30% of the total number “838608”.

More specifically, for example, the physical release control unit 1501 determines the unused area of the container area 121 in response to the fact that the reference counter of any of the data areas 122 of the container area 121 is subtracted to become “0”. Determine if the number exceeds the threshold α. Thereby, it is possible to determine whether or not the number of unused areas of the container area 121 exceeds the threshold value α according to the fact that any of the data areas 122 of the container area 121 is unused.

Further, when the number of unused areas in the container area 121 exceeds the threshold value α, the physical release control unit 1501 indicates that each of the plurality of physical release unit areas in the container area 121 is a data area in use included in the physical release unit area. Calculate the number of 122. In the following description, the number of data areas 122 in use included in the physical release unit area may be referred to as “number of used areas”.

Specifically, for example, the physical release control unit 1501 calculates the number of used areas of each physical release unit area of the container area 121 with reference to the data size management table 312 of the container area 121. The physical release control unit 1501 can specify, for example, which data area 122 is included in which physical release unit area from the slot number of the data size management table 312.

As an example, a plurality of physical release unit areas divided in units of 21 [MB] in ascending order of slot numbers are referred to as physical release unit areas A, B, C, D, and E. In this case, the physical release control unit 1501 calculates the number of used areas of the physical release unit areas A to E for each of the physical release unit areas A to E.

Further, the physical release control unit 1501 transfers data between the physical release unit areas of the container area 121 based on the calculated number of used areas for each physical release unit area. Specifically, for example, first, the physical release control unit 1501 starts with the first and second physical release unit areas from the plurality of physical release unit areas of the container area 121 based on the calculated number of used areas for each physical release unit area. Determine the release unit area.

Here, the first physical release unit area is a physical release unit area that is a data migration source. The second physical release unit area is the physical release unit area to which the data is transferred. More specifically, for example, the physical release control unit 1501 determines the physical release unit area having the smallest number of used areas as the first physical release unit area among the plurality of physical release unit areas of the container area 121. You may. Further, the physical release control unit 1501 may determine the physical release unit area having the largest number of used areas among the plurality of physical release unit areas of the container area 121 as the second physical release unit area.

As an example, it is assumed that the number of used areas of the physical release unit areas A to E is calculated as follows.
Number of used areas in the physical release unit area A "10"
Number of used areas in the physical release unit area B "10"
Number of used areas in the physical release unit area C "8"
Number of used areas "11" in the physical release unit area D
Number of used areas "9" in the physical release unit area E

In this case, the physical release control unit 1501 determines, for example, the physical release unit area C having the smallest number of used areas among the physical release unit areas A to E as the first physical release unit area. Further, the physical release control unit 1501 determines, for example, the physical release unit area D having the largest number of used areas among the physical release unit areas A to E as the second physical release unit area.

Next, the physical release control unit 1501 shifts the data in the used data area 122 included in the first physical release unit area to the unused data area 122 included in the second physical release unit area. The usage status of each data area 122 included in the first and second physical release unit areas is specified from, for example, the data size management table 312.

As a result, data can be transferred between the physical release unit areas in the container area 121. At this time, since the data areas 122 in the container area 121 all have the same data size, data migration can be performed more efficiently than when the data size is not constant.

Further, the physical release control unit 1501 determines that the number of used areas in the first physical release unit area is equal to or less than the number of unused data areas 122 (number of unused areas) included in the second physical release unit area. In addition, data transfer may be performed between the first and second physical release unit areas. The number of unused areas in the physical release unit area can be obtained, for example, by subtracting the number of used areas in the physical release unit area from the total number of data areas 122 included in the physical release unit area. That is, in the physical release control unit 1501, among the plurality of physical release unit areas of the container area 121, the number of unused areas is equal to or greater than the number of used areas of the first physical release unit area, and the number of used areas is larger. The large physical release unit area may be determined as the second physical release unit area. As a result, it is possible to prevent the data area 122 in use from remaining in the first physical release unit area.

However, if the physical release control unit 1501 performs data migration between the first and second physical release unit areas, and as a result, the data area 122 in use remains in the first physical release unit area, a new one is newly used. The second physical release unit region may be determined. At this time, the physical release control unit 1501 selects the remaining physical release unit areas excluding the physical release unit areas determined as the first and second physical release unit areas from the plurality of physical release unit areas of the container area 121. Determine a new second physical release unit region.

As an example, it is assumed that, of the physical release unit areas A to E, the data area 122 in use remains in the physical release unit area C as a result of data migration from the physical release unit area C to the physical release unit area D. .. In this case, the physical release control unit 1501 is, for example, a new second physical unit from the remaining physical release unit areas A, B, E excluding the physical release unit areas C and D in the physical release unit areas A to E. Determine the release unit area. More specifically, for example, the physical release control unit 1501 converts the physical release unit area A having the largest number of used areas among the physical release unit areas A, B, and E into a new second physical release unit area. decide. When there are a plurality of physical release unit areas having the same number of used areas, the physical release control unit 1501 gives priority to, for example, the physical release unit area including the data area 122 having a young slot number.

Further, the physical release control unit 1501 releases the physical storage area corresponding to the first physical release unit area when the data transfer of the data area 122 in use included in the first physical release unit area is completed. do. Here, the physical storage area corresponding to the first physical release unit area is, for example, the physical storage area in the disk 212 allocated to the first physical release unit area.

Specifically, for example, the physical release control unit 1501 is assigned to the first physical release unit area when all the data transfer of the data area 122 in use included in the first physical release unit area is completed. The physical storage area in the disk 212 is cleared to zero. Then, the physical release control unit 1501 updates the control information (not shown) regarding the allocation of the physical storage area of the container area 121. More specifically, for example, the physical release control unit 1501 clears the physical address of the physical storage area associated with the first physical release unit area.

As a result, the first physical release unit area in which the data area 122 in use has disappeared due to data migration between the physical release unit areas in the container area 121 can be physically released.

Further, the physical release control unit 1501 determines new first and second physical release unit areas when all the data transfer of the data area 122 in use included in the first physical release unit area is completed. You may decide. At this time, the physical release unit area in which all the data transfer of the data area 122 in use is completed is excluded from the determination target.

Further, the physical release control unit 1501 has transferred the data in the used data area 122 included in the first physical release unit area to the unused data area 122 included in the second physical release unit area. The block map table 311 and the data size management table 312 are updated accordingly.

Specifically, for example, the physical release control unit 1501 updates the physical storage slot number in the block map table 311 with the physical storage slot number of the migration destination data area 122 for the migrated data. In addition, the physical release control unit 1501 updates the data area usage status and the reference counter of the migration source and migration destination data areas 122 in the data size management table 312. More specifically, for example, the physical release control unit 1501 sets the data area usage status of the migration source data area 122 in the data size-based management table 312 to "unused" and sets the reference counter to "0". Further, the physical release control unit 1501 sets the data area usage status of the migration destination data area 122 in the data size management table 312 to "in use", and sets the reference counter to the value of the reference counter of the migration source data area 122. do.

When the physical release unit area is released, the physical release control unit 1501 may change the usage status of each data area 122 included in the physical release unit area to “empty”. Thereby, for example, when reallocating the physical storage area, the CM211 can determine that the reassignment is necessary if it is confirmed that the first data area 122 is “empty”. If the usage status of all data areas 122 is left as "unused", for example, after confirming that the usage status of all data areas 122 is "unused", the physical storage area is reassigned. It will be a load.

(Example of physical release of physical release unit area)
Next, an example of physical release of the physical release unit area of the container area 121 will be described with reference to FIGS. 16 to 19. Here, an example of physical release of the physical release unit area will be described by taking as an example the container area 121 of the container ID “x” whose usage status has changed significantly due to a change in the data pattern tendency.

FIG. 16 is an explanatory diagram showing an example of the stored contents of the data size management table 312-x. In FIG. 16, the data size management table 312-x has a container ID, container information 501, an unused queue 502, a container management size 503, and a container start address 504.

Container information 501 has records 511-0 to 59. That is, the container area 121 of the container ID “x” has 60 data areas 122 of slot numbers “0” to “59”.

17 to 19 are explanatory views showing an example of physical release of the physical release unit area of the container area 121. In FIG. 17, the usage status of each data area 122 of the container area 121 of the container ID “x” is shown. Here, the upper left is the data area 122 of the slot number “0”, and the data areas 122 are arranged in each line so that the slot numbers are in ascending order from left to right.

For example, in the top row, 10 data areas 122 having slot numbers "0" to "9" are arranged from left to right. Further, the plurality of physical release unit areas of the container area 121 of the container ID "x" are divided into physical release unit areas 1701 to 1703 in units of 20 data areas 122. Further, the threshold value α is set to “α = 20”.

In FIG. 17 (17-1), the usage status of each data area 122 of the container area 121 of the container ID “x” shown in the data size management table 312-x shown in FIG. 16 is shown. In this case, the physical release control unit 1501 refers to the data size management table 312-x and determines whether or not the number of unused areas 121 in the container area 121 of the container ID "x" exceeds the threshold value α.

Here, the number of unused areas in the container area 121 of the container ID “x” is “45”. Therefore, the physical release control unit 1501 determines that the number of unused areas 121 of the container area 121 having the container ID “x” exceeds the threshold value α. Next, the physical release control unit 1501 calculates the number of used areas of the physical release unit areas 1701 to 1703 with reference to the data size management table 312-x.

Here, the number of used areas of the physical release unit area 1701 is "2", the number of used areas of the physical release unit area 1702 is "10", and the number of used areas of the physical release unit area 1703 is "3". In this case, the physical release control unit 1501 determines the physical release unit area 1701 having the smallest number of used areas as the first physical release unit area. Further, the physical release control unit 1501 determines the physical release unit area 1702 having the maximum number of used areas as the second physical release unit area.

Then, the physical release control unit 1501 shifts the data in the used data area 122 included in the physical release unit area 1701 to the unused data area 122 included in the physical release unit area 1702. At this time, the physical release control unit 1501 shifts to the unused data area 122 having a young slot number included in the physical release unit area 1702.

In (17-2) of FIG. 17, the data in the data area 122 of the slot number “1” included in the physical release unit area 1701 is transferred to the data area 122 of the slot number “21” included in the physical release unit area 1702. Has been done.

In (17-3) of FIG. 18, the data in the data area 122 of the slot number “6” included in the physical release unit area 1701 is transferred to the data area 122 of the slot number “23” included in the physical release unit area 1702. Has been done.

When all the data transfer of the data area 122 in use included in the physical release unit area 1701 is completed, the physical release control unit 1501 determines new first and second physical release unit areas. Here, the number of used areas of the physical release unit area 1702 after the completion of data migration is "12", and the number of used areas of the physical release unit area 1703 is "3".

In this case, the physical release control unit 1501 determines the physical release unit area 1703, which has the smallest number of used areas, as the first physical release unit area. Further, the physical release control unit 1501 determines the physical release unit area 1702 having the maximum number of used areas as the second physical release unit area. The physical release unit area 1701 for which data migration has been completed is excluded from the determination target.

Then, the physical release control unit 1501 shifts the data in the used data area 122 included in the physical release unit area 1703 to the unused data area 122 included in the physical release unit area 1702. At this time, the physical release control unit 1501 shifts to the unused data area 122 having a young slot number included in the physical release unit area 1702.

In (17-4) of FIG. 18, the data in the data area 122 of the slot number “48” included in the physical release unit area 1703 is transferred to the data area 122 of the slot number “24” included in the physical release unit area 1702. Has been done.

Similarly, the data in the data area 122 of the slot number "52,55" included in the physical release unit area 1703 is transferred to the data area 122 of the slot number "26,28" included in the physical release unit area 1702. Then, the state shown in (17-5) of FIG. 19 is obtained.

Then, the physical release control unit 1501 releases the physical storage area corresponding to the physical release unit area in which the data transfer of the data area 122 in use is completed. In (17-6) of FIG. 19, the physical storage area corresponding to the physical release unit area 1701, 1703 is released.

As a result, garbage collection of the container area 121 of the container ID "x" in which the unused data area 122 has increased due to the change in the data pattern tendency is performed, and the usage amount of the physical storage area is reduced to one-third. Can be done. The physically released area can be used for other container areas 121 and the like, and the usage efficiency of the disk 212 can be improved.

Further, since the data areas 122 in the container area 121 all have the same data size, data migration can be performed more efficiently than when the data size is not constant. For example, if the data size is not constant, the data in use will be moved, whereas in the examples shown in FIGS. 17 to 19, the number of writes can be reduced to five.

(Various processing procedures of CM211)
Next, various processing procedures of CM211 according to the second embodiment will be described. First, the writing procedure at the time of receiving Write I / O of CM211 according to the second embodiment will be described. However, except for the specific processing procedure of the reference counter addition / subtraction processing in step S1113 shown in FIG. 11, the procedure is the same as the writing I / O reception processing procedure of CM211 according to the first embodiment. Therefore, a specific processing procedure of the reference counter addition / subtraction processing different from that of the CM211 according to the first embodiment will be described.

FIG. 20 is a flowchart showing an example of the reference counter addition / subtraction processing procedure according to the second embodiment. The CM211 adds 1 to the reference counter of the data area 122 in which the current compression division Write data wdsc is written (step S2001). Further, the CM 211 subtracts 1 from the reference counter of the data area 122 that was used before writing the compression division write data this time (step S2002). Then, the CM211 determines whether or not there is a data area 122 in which the reference counter has become 0 by updating the reference counter this time (step S2003).

When there is a data area 122 in which the reference counter has become 0 (step S2003: there is a data area in which the reference counter has become 0), the CM 211 assumes that the target data area 122 is unused and puts the target data area in the unused queue. 122 is connected (step S2004).

Next, CM211 determines whether or not the number of unused areas of the target container area 121 exceeds the threshold value α (step S2005). The target container area 121 is a container area 121 in which the target data area 122 is connected to the unused queue in step S2004.

When the number of unused areas of the target container area 121 exceeds the threshold value α (step S2005: the threshold value is exceeded), the CM 211 calls the garbage collection process (step S2006). The specific processing procedure of the garbage collection processing will be described later with reference to FIGS. 21 and 22.

After the processing of step S2006 is completed, or when there is no data area 122 whose reference counter has become 0 (step S2003: no data area whose reference counter has become 0), or the number of unused areas of the target container area 121. When is equal to or less than the threshold value α (step S2005: equal to or less than the threshold value), CM211 ends the reference counter addition / subtraction process.

As a result, it is possible to determine whether or not the garbage collection of the container area 121 needs to be performed at the timing when the unused data area 122 of the container area 121 increases.

21 and 22 are flowcharts showing an example of the garbage collection processing procedure according to the second embodiment. In the flowchart of FIG. 21, first, the CM 211 calculates the number of used areas of the physical release unit area of the target container area 121 by referring to the data size management table 312 of the target container area 121 (step S2101).

Next, the CM 211 connects the physical release unit area of the target container area 121 to the garbage block link in ascending order of the calculated number of used areas (step S2102). Then, CM211 determines the physical release unit area at the beginning of the garbage block link as the migration source area (step S2103). The migration source region corresponds to the first physical release unit region described above.

Each area connected to the garbage block link has prev information and next information, and is in a state where the immediately preceding area and the next area can be specified.

Next, CM211 determines the physical release unit area at the end of the garbage block link as the migration destination area (step S2104). The migration destination region corresponds to the second physical release unit region described above. Next, the CM 211 determines whether or not there is a data area 122 in use in the determined migration source area (step S2105).

Here, when there is a data area 122 in use in the migration source area (step S2105: Yes), the CM211 determines whether or not there is an unused data area 122 in the determined migration destination area (step S2106). .. Here, when there is an unused data area 122 in the migration destination area (step S2106: Yes), the CM 211 shifts to step S2201 shown in FIG.

Further, in step S2105, when there is no data area 122 in use in the migration source area (step S2105: No), the CM211 has a physical release unit area next to the migration source area based on the next information of the migration source area. It is determined whether or not (step S2107).

Here, when there is a physical release unit area next (step S2107: Yes), CM211 changes the next physical release unit area to the migration source area (step S2108), and proceeds to step S2105. On the other hand, if there is no physical release unit area next (step S2107: No), CM211 shifts to step S2207 shown in FIG.

Further, in step S2106, when there is no unused data area 122 in the migration destination area (step S2106: No), does CM211 have a physical release unit area immediately before the migration destination area based on the prev information of the migration destination area? It is determined whether or not (step S2109).

Here, if there is a physical release unit area immediately before (step S2109: Yes), the CM211 changes the immediately preceding physical release unit area to the migration destination area (step S2110), and proceeds to step S2106. On the other hand, when there is no physical release unit area immediately before (step S2109: No), CM211 shifts to step S2207 shown in FIG.

In the flowchart of FIG. 22, first, CM211 determines whether or not the migration source region and the migration destination region are different regions (step S2201). Here, when the migration source area and the migration destination area are different areas (step S2201: Yes), the CM211 determines whether or not the number of used areas of the migration source area is equal to or less than the number of used areas of the migration destination area (step S2201: Yes). Step S2202).

Here, when the number of used areas of the migration source area is equal to or less than the number of used areas of the migration destination area (step S2202: Yes), the CM211 transfers the data of the data area 122 in use included in the migration source area to the migration destination area. It is copied to the unused data area 122 included in (step S2203).

Then, CM211 updates the block map table 311 and the data size management table 312 (step S2204). Next, CM211 determines whether or not there is a data area 122 in use in the migration source area (step S2205).

Here, when there is no data area 122 in use in the migration source area (step S2205: No), CM211 shifts to step S2107 shown in FIG. On the other hand, when there is a data area 122 in use in the migration source area (step S2205: Yes), the CM211 determines whether or not there is an unused data area 122 in the migration destination area (step S2206).

Here, when there is no unused data area 122 in the migration destination area (step S2206: No), CM211 shifts to step S2109 shown in FIG. On the other hand, when there is an unused data area 122 in the migration destination area (step S2206: Yes), CM211 returns to step S2203.

Further, in step S2201, when the migration source area and the migration destination area are the same area (step S2201: No), or in step S2202, the number of used areas of the migration source area is larger than the number of used areas of the migration destination area (step S2201: No). Step S2202: No), CM211 shifts to step S2207. As a result, it is possible to prevent data migration when the number of used areas in the migration source area becomes larger than the number of used areas in the migration destination area.

Then, the CM 211 physically releases the physical release unit area in which all the data areas 122 are unused (step S2207), and ends a series of processes according to this flowchart. As a result, garbage collection can be performed on a container-by-container basis to physically release areas that are no longer used due to changes in data pattern trends.

As described above, the CM 211 according to the second embodiment determines whether or not the number of unused areas in the container area 121 exceeds the threshold value α. Specifically, for example, in the CM 211, the number of unused areas in the container area 121 sets the threshold value α according to the fact that the reference counter in any data area 122 in the container area 121 is subtracted to become “0”. Determine if it has been exceeded.

Thereby, it is possible to determine whether or not the number of unused areas of the container area 121 exceeds the threshold value α according to the fact that any of the data areas 122 of the container area 121 is unused. In other words, it is possible to determine whether or not the garbage collection of the container area 121 needs to be performed at the timing when the unused data area 122 of the container area 121 increases.

Further, in the CM 211, when the number of unused areas of the container area 121 exceeds the threshold value α, the number of used areas of the data area in use included in the physical release unit area for each of the plurality of physical release unit areas of the container area 121. Is calculated. Further, the CM211 obtains the data of the data area 122 in use included in the first physical release unit area among the plurality of physical release unit areas, based on the calculated number of used areas for each physical release unit area. It shifts to the unused data area 122 included in the physical release unit area of 2. Then, the CM 211 releases the physical storage area corresponding to the first physical release unit area when the data transfer of the data area 122 in use included in the first physical release unit area is completed.

As a result, when the data pattern tendency changes and the number of unused areas in the container area 121 exceeds the threshold value α, data is transferred between the physical release unit areas in the container area 121 to create a cohesive unused area. Physical release (garbage collection) can be performed.

Further, the CM 211 determines the physical release unit area having the smallest number of used areas as the first physical release unit area among the plurality of physical release unit areas of the container area 121, and the physical release unit having the largest number of used areas. The region can be determined as the second physical release unit region. As a result, it is possible to efficiently create a large unused area in the container area 121.

Further, the CM 211 includes data in use included in the first physical release unit area when the number of used areas in the first physical release unit area is equal to or less than the number of unused areas in the second physical release unit area. The data in the area 122 can be transferred to the unused data area 122 included in the second physical release unit area. As a result, it is possible to prevent the data area 122 in use from remaining in the first physical release unit area because the data cannot be completely transferred to the second physical release unit area.

Based on these facts, according to the CM 211 according to the second embodiment, it is possible to carry out garbage collection of the container area 121 in which the unused area has increased due to the change in the data pattern tendency, and improve the usage efficiency of the disk 212. can. Further, since the data areas 122 in the container area 121 all have the same data size, data migration can be performed more efficiently than when the data size is not constant. Therefore, garbage collection can be performed while suppressing the influence on the life of the SSD or the like used for the disk 212.

The storage control method described in the present embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This storage control program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM (Compact Disc-Read Only Memory), or a DVD (Digital Versaille Disk), and is read from the recording medium by the computer. Is executed by. Further, this storage control program may be distributed via a network such as the Internet.

The following additional notes are further disclosed with respect to the above-described embodiment.

(Appendix 1) In the storage having a storage area including a plurality of data areas which are integral multiples of the size of the access unit of the storage, each data of the plurality of data areas of the first size which is the integral multiple allocated from the storage area. The first management information having information capable of specifying the physical address of the area and information indicating the usage status of each data area, and the integral multiple of the storage area allocated, which is different from the first size. The second management information having the information that can specify the physical address of each data area of the plurality of data areas of the second size and the information indicating the usage status of each data area, and the data in the storage area. A storage unit having logical address information in which the logical address of the data is associated with information that can identify the physical address of the written data area, and
Depending on the acceptance of the write data of a predetermined size and the logical address of the write data, the size of the compressed data obtained by compressing the write data or more from the first management information and the second management information The management information corresponding to the size of is specified, the compressed data is written to the unused data area among the plurality of data areas specified from the specified management information, and the physical of the unused data area is written. A control unit that stores the information in which the address can be specified and the logical address of the write data in the logical address information in association with each other.
A storage control device characterized by having.

(Appendix 2) The storage area is a storage area accessed by the head.
The plurality of data areas of an integral multiple are continuous areas, and the plurality of data areas are continuous areas.
The control unit
If there are a plurality of unused data areas among the plurality of data areas specified from the specified management information, the compressed data is written to the unused data area at the head.
The storage control device according to Appendix 1, wherein the storage control device is characterized by the above.

(Appendix 3) The information that can identify the physical addresses of the plurality of data areas in the first management information includes the first physical address of the plurality of data areas and the first data in each data area of the plurality of data areas. The order from the area and the first size,
The information that can identify the physical addresses of the plurality of data areas in the second management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the second size,
The logical address information includes management information to which the data area in which the data is written belongs, the order of the data area from the beginning of a plurality of data areas including the data area, and the logical address of the data. Is associated with
The control unit
In response to receiving the logical address of the read range of a predetermined size, the management information and the order corresponding to the logical address of the read range are specified by referring to the logical address information.
Data in which data corresponding to the logical address of the read range is written based on the physical address at the beginning of a plurality of data areas in the specified management information, the specified order, and the size corresponding to the management information. Calculate the physical address of the area and
Read data from the physical address in the storage area,
The storage control device according to Appendix 1 or 2, wherein the storage control device is described.

(Appendix 4) The information that can identify the physical addresses of the plurality of data areas in the first management information includes the first physical address of the plurality of data areas and the first data in each data area of the plurality of data areas. The order from the area and the first size,
The information that can identify the physical addresses of the plurality of data areas in the second management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the second size,
The logical address information includes management information to which the data area in which the data is written belongs, the order of the data area from the beginning of a plurality of data areas including the data area, and the logical address of the data. Is associated with
The control unit
When the management information corresponding to the logical address of the write data in the logical address information and the specified management information are the same, the logical address information among the plurality of data areas specified from the specified management information. The compressed data is written in the data area indicated in the order corresponding to the logical address of the write data in the above.
The storage control device according to any one of Appendix 1 to 3, wherein the storage control device is characterized by the above.

(Appendix 5) Each data area of the plurality of data areas in the first management information has data that does not overlap with each other.
The first management information further has a number of logical addresses that refer to each data area corresponding to each data area of the plurality of data areas in the first management information.
Each data area of the plurality of data areas in the second management information has data that does not overlap with each other.
The second management information further has a number of logical addresses that refer to each data area corresponding to each data area of the plurality of data areas in the second management information.
The control unit
Data in which the management information corresponding to the logical address of the write data in the logical address information and the specified management information are the same and are shown in the order corresponding to the logical address of the write data in the logical address information. When the number of logical addresses that refer to the area is 1, the data area indicated in the order corresponding to the logical address of the write data in the logical address information among the plurality of data areas specified from the specified management information. Write the compressed data to
The storage control device according to Appendix 4, wherein the storage control device is characterized by the above.

(Appendix 6) The control unit is
When the number of unused data areas among the plurality of data areas exceeds the threshold value, each of the plurality of physical release unit areas in which the plurality of data areas are divided into units that can be physically released is assigned to the physical release unit area. Calculate the number of used areas of the included data area in use,
Based on the calculated number of used areas for each physical release unit area, the data of the data area in use included in the first physical release unit area among the plurality of physical release unit areas is released to the second physical release unit area. Migrate to the unused data area contained in the unit area,
When the data transfer of the data area in use included in the first physical release unit area is completed, the physical storage area corresponding to the first physical release unit area is released.
The storage control device according to any one of Supplementary note 1 to 5, wherein the storage control device is described.

(Appendix 7) The first physical release unit area is a physical release unit area having the smallest number of used areas among the plurality of physical release unit areas.
The second physical release unit area is a physical release unit area having the largest number of used areas among the plurality of physical release unit areas.
The storage control device according to Appendix 6, wherein the storage control device is characterized by the above.

(Appendix 8) The control unit is
Use included in the first physical release unit area when the number of used areas of the first physical release unit area is equal to or less than the number of unused data areas included in the second physical release unit area. The data in the data area inside is transferred to the unused data area included in the second physical release unit area.
The storage control device according to Appendix 6 or 7, wherein the storage control device is described.

(Appendix 9) The control unit is
The management information and the logic in response to the transfer of the data in the used data area included in the first physical release unit area to the unused data area included in the second physical release unit area. Update address information,
The storage control device according to any one of Supplementary note 6 to 8, wherein the storage control device is characterized by the above.

(Appendix 10) To the computer
The storage in response to the reception of write data of a predetermined size and the logical address of the write data for the storage having a storage area including a plurality of data areas that are integral multiples of the size of the access unit of the storage. The first management information having information that can specify the physical address of each data area of the plurality of data areas of the first size of the integral multiple allocated from the area and information indicating the usage status of each data area, and Information that can specify the physical address of each data area of a plurality of data areas of a second size that is an integral multiple of the storage area and is different from the first size, and the usage status of each data area. From the second management information having the information indicating the above, the management information corresponding to the size equal to or larger than the size of the compressed data obtained by compressing the written data is specified.
The compressed data is written to the unused data area among the plurality of data areas specified from the specified management information, and the information capable of specifying the physical address of the unused data area and the written data The physical address of the data area in which the data is written in the storage area is stored in the logical address information in which the logical address of the data is associated with the identifiable information by associating with the logical address.
A storage control program characterized by executing processing.

(Appendix 11) The information that can identify the physical addresses of the plurality of data areas in the first management information includes the first physical address of the plurality of data areas and the first data in each data area of the plurality of data areas. The order from the area and the first size,
The information that can identify the physical addresses of the plurality of data areas in the second management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the second size,
The logical address information includes management information to which the data area in which the data is written belongs, the order of the data area from the beginning of a plurality of data areas including the data area, and the logical address of the data. Is associated with
On the computer
In response to receiving the logical address of the read range of a predetermined size, the management information and the order corresponding to the logical address of the read range are specified by referring to the logical address information.
Data in which data corresponding to the logical address of the read range is written based on the physical address at the beginning of a plurality of data areas in the specified management information, the specified order, and the size corresponding to the management information. Calculate the physical address of the area and
Read data from the physical address in the storage area,
The storage control program according to Appendix 10, wherein the process is executed.

101 Storage control device 102 Storage 110 Management information 111 Logical address information 121 Container area 122 Data area 300 Control unit 301 Reception unit 302 Container area determination unit 303 Writing unit 304 Specific unit 305 Reading unit 310 Storage unit 311 Block map table 312 By data size Management table 1501 Physical release control unit 1701, 1702, 1703 Physical release unit area

Claims (8)

In the storage having a storage area including a plurality of data areas that are integral multiples of the size of the access unit of the storage, the physical address of each data area of the plurality of data areas of the first size that is the integral multiples allocated from the storage area. A first management information having information that can identify the data and information indicating the usage status of each data area, and a second size that is an integral multiple of the storage area and is different from the first size. Second management information having information that can identify the physical address of each data area of the plurality of data areas and information indicating the usage status of each data area, and a data area in which the data in the storage area is written. A storage unit having logical address information in which the logical address of the data is associated with information that can identify the physical address of the data.
Depending on the acceptance of the write data of a predetermined size and the logical address of the write data, the size of the compressed data obtained by compressing the write data or more from the first management information and the second management information The management information corresponding to the size of is specified, the compressed data is written to the unused data area among the plurality of data areas specified from the specified management information, and the physical of the unused data area is written. The information whose address can be specified is associated with the logical address of the written data and stored in the logical address information.
When the number of unused data areas among the plurality of data areas exceeds the threshold value, each of the plurality of physical release unit areas in which the plurality of data areas are divided into units that can be physically released is assigned to the physical release unit area. Calculate the number of used areas of the included data area in use,
Based on the calculated number of used areas for each physical release unit area, the data of the data area in use included in the first physical release unit area among the plurality of physical release unit areas is released to the second physical release unit area. Migrate to the unused data area contained in the unit area,
A control unit that releases the physical storage area corresponding to the first physical release unit area when the data transfer of the data area in use included in the first physical release unit area is completed.
A storage control device characterized by having. The storage area is a storage area accessed by the head.
The plurality of data areas of an integral multiple are continuous areas, and the plurality of data areas are continuous areas.
The control unit
If there are a plurality of unused data areas among the plurality of data areas specified from the specified management information, the compressed data is written to the unused data area at the head.
The storage control device according to claim 1. The plurality of data areas of an integral multiple are continuous areas, and the plurality of data areas are continuous areas.
The information that can identify the physical addresses of the plurality of data areas in the first management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the first size
The information that can identify the physical addresses of the plurality of data areas in the second management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the second size,
The logical address information includes management information to which the data area in which the data is written belongs, the order of the data area from the beginning of a plurality of data areas including the data area, and the logical address of the data. Is associated with
The control unit
In response to receiving the logical address of the read range of a predetermined size, the management information and the order corresponding to the logical address of the read range are specified by referring to the logical address information.
Data in which data corresponding to the logical address of the read range is written based on the physical address at the beginning of a plurality of data areas in the specified management information, the specified order, and the size corresponding to the management information. Calculate the physical address of the area and
Read data from the physical address in the storage area,
The storage control device according to claim 1 or 2. The plurality of data areas of an integral multiple are continuous areas, and the plurality of data areas are continuous areas.
The information that can identify the physical addresses of the plurality of data areas in the first management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the first size
The information that can identify the physical addresses of the plurality of data areas in the second management information is the order from the first physical address of the plurality of data areas and the first data area in each data area of the plurality of data areas. And the second size,
The logical address information includes management information to which the data area in which the data is written belongs, the order of the data area from the beginning of a plurality of data areas including the data area, and the logical address of the data. Is associated with
The control unit
When the management information corresponding to the logical address of the write data in the logical address information and the specified management information are the same, the logical address information among the plurality of data areas specified from the specified management information. The compressed data is written in the data area indicated in the order corresponding to the logical address of the write data in the above.
The storage control device according to any one of claims 1 to 3. Each data area of the plurality of data areas in the first management information has data that does not overlap with each other.
The first management information further has a number of logical addresses that refer to each data area corresponding to each data area of the plurality of data areas in the first management information.
Each data area of the plurality of data areas in the second management information has data that does not overlap with each other.
The second management information further has a number of logical addresses that refer to each data area corresponding to each data area of the plurality of data areas in the second management information.
The control unit
Data in which the management information corresponding to the logical address of the write data in the logical address information and the specified management information are the same and are shown in the order corresponding to the logical address of the write data in the logical address information. When the number of logical addresses that refer to the area is 1, the data area indicated in the order corresponding to the logical address of the write data in the logical address information among the plurality of data areas specified from the specified management information. Write the compressed data to
The storage control device according to claim 4. The first physical release unit area is a physical release unit area having the smallest number of used areas among the plurality of physical release unit areas.
The second physical release unit area is a physical release unit area having the largest number of used areas among the plurality of physical release unit areas.
The storage control device according to claim 1. The control unit
Use included in the first physical release unit area when the number of used areas of the first physical release unit area is equal to or less than the number of unused data areas included in the second physical release unit area. The data in the data area inside is transferred to the unused data area included in the second physical release unit area.
The storage control device according to claim 6. On the computer
The storage in response to the reception of write data of a predetermined size and the logical address of the write data for the storage having a storage area including a plurality of data areas that are integral multiples of the size of the access unit of the storage. The first management information having information that can specify the physical address of each data area of the plurality of data areas of the first size of the integral multiple allocated from the area and information indicating the usage status of each data area, and Information that can specify the physical address of each data area of a plurality of data areas of a second size that is an integral multiple of the storage area and is different from the first size, and the usage status of each data area. From the second management information having the information indicating the above, the management information corresponding to the size equal to or larger than the size of the compressed data obtained by compressing the written data is specified.
The compressed data is written to the unused data area among the plurality of data areas specified from the specified management information, and the information capable of specifying the physical address of the unused data area and the written data By associating with the logical address, the physical address of the data area in which the data is written in the storage area is stored in the logical address information in which the logical address of the data is associated with the identifiable information.
When the number of unused data areas among the plurality of data areas exceeds the threshold value, each of the plurality of physical release unit areas in which the plurality of data areas are divided into units that can be physically released is assigned to the physical release unit area. Calculate the number of used areas of the included data area in use,
Based on the calculated number of used areas for each physical release unit area, the data of the data area in use included in the first physical release unit area among the plurality of physical release unit areas is released to the second physical release unit area. Migrate to the unused data area contained in the unit area,
When the data transfer of the data area in use included in the first physical release unit area is completed, the physical storage area corresponding to the first physical release unit area is released.
A storage control program characterized by executing processing.

Images