JP7722075B2 - Storage device, storage system, and backup method - Google Patents

Description

The present invention relates to a storage device, a storage system, and a backup method.

Block storage, which stores data in blocks, reduces the amount of data transfer by transferring updated blocks of data when backing up data to other storage. The backup destination storage holds the transferred data in cache memory and backs up the transferred data by writing the data from the cache memory to the storage medium. In such storage systems, if the amount of data transferred for backup increases, the write process can become a rate-limiting factor in slow storage media, which can slow down the backup processing speed. This is particularly true when the blocks on the backup destination storage medium are not contiguous. Patent Document 1, for example, discloses a technology for reducing the reduction in processing speed when backing up data between block storages.

The storage system of Patent Document 1 writes data directly to the storage medium when the blocks to which data is written are contiguous in the backup destination storage device. Furthermore, when the blocks are not contiguous, the storage system of Patent Document 1 writes the data to a high-speed storage medium first, and then writes the data to the storage medium.

JP 2014-041471 A

However, the storage system described in Patent Document 1 may not be able to determine block continuity if the backup destination storage device uses thin provisioning functionality.

The present invention aims to provide a storage device or the like that can prevent a decrease in backup processing speed.

To solve the above problem, the storage device of the present invention comprises an information management means for managing mapping information obtained from the backup destination storage device; a data management means for managing, on a block-by-block basis, whether data in the storage area has been updated since the previous backup; a determination means for determining, based on the mapping information, whether the backup destination of the updated block data is a contiguous block in the backup destination storage device; and a transmission control means for controlling the transmission of an instruction to write to the backup destination storage medium and the backup data to the backup destination storage device if the backup destination is a contiguous block, and for controlling the transmission of an instruction to write to the extended cache memory and the backup data to the backup destination storage device if the backup destination is a distant block.

The backup method of the present invention manages mapping information obtained from the backup destination storage device, manages whether data in the storage area has been updated since the previous backup on a block-by-block basis, determines based on the mapping information whether the backup destination for the updated block data is a contiguous block on the backup destination storage device, and, if the blocks are contiguous, controls the transmission of an instruction to write to the backup destination block and the backup data to the backup destination storage device. If the blocks are distant, controls the transmission of an instruction to write to the extended cache memory and the backup data to the backup destination storage device.

This invention makes it possible to prevent a decrease in backup processing speed.

1 is a diagram illustrating an outline of a configuration of a first exemplary embodiment of the present invention. FIG. 1 illustrates an example of a configuration of a storage apparatus according to a first embodiment of the present invention. 10 illustrates an example of a differential bitmap according to the first embodiment of the present invention. FIG. 1 illustrates an example of a configuration of a storage apparatus according to a first embodiment of the present invention. FIG. 4 is a diagram illustrating an example of mapping information according to the first exemplary embodiment of the present invention. FIG. 4 illustrates an example of an operation flow of the storage apparatus according to the first embodiment of the present invention. FIG. 4 illustrates an example of an operation flow of the storage apparatus according to the first embodiment of the present invention. FIG. 4 illustrates an example of an operation flow of the storage apparatus according to the first embodiment of the present invention. FIG. 10 is a diagram illustrating an outline of the configuration of a second embodiment of the present invention. FIG. 10 is a diagram illustrating an example of an operation flow of a storage apparatus according to a second embodiment of the present invention.

(First embodiment)
A first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing an outline of the configuration of an information processing system according to this embodiment. The information processing system includes a server 10, a storage device 20, and a storage device 30. The storage device 20 is a backup source storage device when backing up data in the information processing system. The storage device 30 is a backup destination storage device when backing up data in the information processing system. The server 10 and the storage device 20 are connected via a network. The storage devices 20 and 30 are also connected via a network.

In this embodiment of the information processing system, storage device 20 and storage device 30 each manage storage areas in block units. Server 10 writes data to storage device 20 and reads data from storage device 20. Storage device 20 uses a replication function to back up stored data to storage device 30. Storage device 20 stores differential bitmap data that manages whether data has been updated in block units since the previous backup. Storage device 20 reduces the amount of data transferred by transferring only the data of updated blocks to storage device 30 as backup data.

The storage device 30, which is the data backup destination, has a thin provisioning function. Thin provisioning is a technology that virtualizes storage capacity in a storage device. With thin provisioning, a virtual volume is allocated when a request to secure space is made, and a physical area is allocated when actual writing is performed.

The server 10 is an information processing device that performs processing according to the application. The server 10 reads data required for processing executed on the server 10 from the storage device 20. The server 10 also writes data generated by executing the processing to the storage device 20.

The configuration of the storage device 20 will now be described. Figure 2 is a diagram showing an example of the configuration of the storage device 20. The storage device 20 includes a control unit 21, an acquisition unit 22, a data transmission unit 23, and a storage medium 100.

The control unit 21 manages the storage area of the storage medium 100 and controls the backup of data stored on the storage medium 100 to the storage device 30. The control unit 21 further includes a data management unit 24, an information management unit 25, an information storage unit 26, a determination unit 27, and a transmission control unit 28.

The data management unit 24 manages for each block whether data has been updated since the previous backup was performed. The data management unit 24 stores data indicating whether each block has been updated as a differential bitmap. The differential bitmap is a data table that divides the storage area into blocks and indicates for each block whether data has been updated since the previous backup was performed.

Figure 3 is a diagram that shows a schematic image of a differential bitmap. A differential bitmap is a table that divides a storage area into blocks. The differential bitmap stores "1" for blocks that have been updated and "0" for blocks that have not been updated. The blocks are arranged in order according to the area of the logical volume to be backed up. A logical volume is a unit for dividing storage area in block storage. From the information in the differential bitmap, the amount of data difference since the previous backup can be obtained as the number of blocks that have been updated. Furthermore, from the information in the differential bitmap, information on the blocks that will be the target of the next backup can be obtained.

The information management unit 25 manages mapping information for the storage device 30. The data management unit 24 acquires mapping information for the thin provisioning function from the storage device 30 via the acquisition unit 22. The mapping information associates logical volumes with the physical areas of the storage media allocated to each logical volume.

The information management unit 25 manages a data table showing the correspondence between each block of the storage medium 100 and the backup destination block in the storage device 30. The information management unit 25 generates a data table by associating each block of the storage medium 100 with the backup destination block in the storage device 30 based on mapping information. The backup destination block in the storage device 30 is notified to the storage device 20 from the storage device 30 using a logical volume when data is backed up for the first time.

The information storage unit 26 stores data necessary for data backup processing. The information storage unit 26 stores differential bitmaps, mapping information, and data tables that indicate the correspondence between each block on the storage medium 100 and the backup destination block on the storage device 30.

The determination unit 27 determines whether the blocks on the storage device 30, which is the backup destination of the data, are contiguous. The determination unit 27 determines whether the blocks on the backup destination are contiguous by referencing the mapping information of the thin provisioning function of the backup destination storage device 30 acquired by the information management unit 25 and determining whether the blocks on the backup destination are contiguous.

The transmission control unit 28 uses the replication function of the block storage to control the backup of data to the storage device 30 using asynchronous differential copying. The transmission control unit 28 references the differential bitmap and transfers data of blocks that have been updated since the previous backup was performed to the storage device 30 as backup data. Backup data is data that is to be backed up when backing up data from the storage device 20, which is the backup source, to the storage device 30, which is the backup destination.

The transmission control unit 28 determines the write destination of the data to be backed up based on the determination result of the determination unit 27. If the determination result shows that the blocks of the backup destination are contiguous, the transmission control unit 28 controls the data transmission unit 23 and the storage medium 100 to send an instruction to write the data to the storage medium and the backup data to the backup destination storage device 30. If the determination result shows that the blocks of the backup destination are not contiguous, i.e., if the blocks of the backup destination are separated, the transmission control unit 28 controls the data transmission unit 23 and the storage medium 100 to send an instruction to write to the extended cache memory and the backup data to the storage device 30.

The acquisition unit 22 acquires data used for each process in the storage device 20. The acquisition unit 22 acquires a data write request and the data to be written from the server 10. The acquisition unit 22 acquires a data read request from the server 10. The acquisition unit 22 also acquires mapping information for the storage device 30 from the storage device 30.

The data transmission unit 23 transmits information about the data write destination and backup data to the storage device 30. The data transmission unit 23 transmits the data requested for reading to the server 10.

The processes in the data management unit 24, information management unit 25, determination unit 27, transmission control unit 28, acquisition unit 22, and data transmission unit 23 of the control unit 21 are performed, for example, by executing a computer program on a CPU (Central Processing Unit). The processes in the data management unit 24, information management unit 25, determination unit 27, transmission control unit 28, acquisition unit 22, and data transmission unit 23 of the control unit 21 may be performed, for example, by a semiconductor device such as an FPGA (Field Programmable Gate Array). The information storage unit 26 is configured, for example, using a non-volatile semiconductor storage device.

The storage medium 100 is configured using, for example, a hard disk drive. However, the storage medium 100 is not limited to a hard disk drive. The storage medium 100 may be configured with multiple hard disk drives. Furthermore, the storage medium 100 may be provided in a housing separate from the storage device 20, as long as it is configured to be controlled by the storage device 20.

The configuration of storage device 30 will now be described. Figure 4 is a diagram showing an example of the configuration of storage device 30. Storage device 30 comprises a control unit 31, an output unit 32, a data receiving unit 33, a storage medium 110, and an extended cache memory 120. Storage device 30 is a storage device that manages storage areas using a thin provisioning function.

The control unit 31 manages the storage area of the storage medium 110. It also controls the writing of data to the storage medium 110 and the extended cache memory 120, and the writing of data stored in the extended cache memory to the storage device 30. The control unit 31 further includes an area management unit 34, a mapping information storage unit 35, and a write control unit 36.

The area management unit 34 manages the data stored on the storage medium 110 on a block-by-block basis. The area management unit 34 manages the storage area using the thin provisioning function. The area management unit 34 sends mapping information to the storage device 20 via the output unit 32.

Figure 5 shows an example of mapping information used by the area management unit 34 to manage storage areas. The mapping information consists of a logical volume number, logical volume block number, disk number, and disk block number. The logical volume number is the identification number of each logical volume. The logical volume block number is the identification number of each block within the logical volume. The disk number and disk block number are the identification numbers of the disk and block within the disk corresponding to each block in the logical volume. In storage devices that use the thin provisioning function, logical volume data is often distributed across multiple storage media. Therefore, the area management unit 34 uses mapping information such as that shown in Figure 5 to associate and manage logical volumes with corresponding storage media.

The mapping information storage unit 35 stores the mapping information.

The write control unit 36 controls the writing of data to the extended cache memory 120 and the storage medium 110. When a command to write to the extended cache memory is issued, the write control unit 36 controls the writing of data received from the storage device 20 to the extended cache memory 120 via the data receiving unit 33. Furthermore, when the access frequency to the storage medium 110 is below a certain level, the write control unit 36 controls the writing of data written to the extended cache memory 120 to the storage medium 110. The access frequency standard is set appropriately using, for example, the number of accesses to the storage medium 110 per unit time. When data from the extended cache memory 120 is written to the storage medium 110, the write control unit 36 erases the backup data from the extended cache memory 120 after it has been written to the storage medium 110. When a command to write to the storage medium is issued, the write control unit 36 controls the writing of data received from the storage device 20 to the storage medium 110 via the data receiving unit 33.

The output unit 32 sends the mapping information to the storage device 20. When a data read request is received, the output unit 32 outputs the specified data on the storage medium 110 to the storage device 20 or the server 10.

The data receiving unit 33 receives a data write request, a data write destination, and backup data from the storage device 20. The data receiving unit 33 receives a request for mapping information from the storage device 20.

The processes in the area management unit 34 of the control unit 31, the write control unit 36, the output unit 32, and the data receiving unit 33 are performed, for example, by executing a computer program on a CPU. The processes in the area management unit 34, the write control unit 36, the output unit 32, and the data receiving unit 33 may also be performed by a semiconductor device such as an FPGA. The mapping information storage unit is configured, for example, using a non-volatile semiconductor storage device.

The storage medium 110 is configured using, for example, a hard disk drive. The storage medium 110 may also be configured using multiple hard disk drives. Furthermore, the storage medium 110 may be provided in a housing separate from the storage device 30, as long as it is configured to be controlled by the storage device 30.

The extended cache memory 120 is configured using, for example, a non-volatile semiconductor memory device.

The operation of the information processing system of this embodiment will be explained.

First, we will explain the operation when the server 10 writes data to the storage medium 100 of the storage device 20. The server 10 sends a data write instruction and the data to be written to the storage device 20. Upon receiving the write instruction, the data management unit 24 controls the storage medium 100 so that the received data is written to the storage medium 100.

When data is written to the storage medium 100, the data management unit 24 updates the differential bitmap information stored in the information storage unit 26 for the block to which data has been written. The data management unit 24 updates the data of the block to which data has been written to "1".

Next, we will explain the operation of storage device 20 acquiring mapping information from storage device 30. Figure 6 is a diagram showing an example of the operation flow when storage device 20 acquires mapping information.

The information management unit 25 checks whether a predetermined time has passed since the previous acquisition of mapping information. The predetermined time is set in advance based on, for example, the update frequency of mapping information in the storage device 30. If the predetermined time has not passed since the previous acquisition of mapping information (No in step S11), the information management unit 25 waits until the predetermined time has passed. If the predetermined time has passed since the previous acquisition (Yes in step S11), the information management unit 25 acquires mapping information from the storage device 30 via the acquisition unit 22 (step S12). After acquiring the mapping information, the information management unit 25 updates the mapping information stored in the information storage unit 26 with the acquired data (step S13).

Next, we will explain the process by which storage device 20 backs up data stored on storage medium 100 to storage device 30. Figure 7 is a diagram showing an example of the operational flow of storage device 20 when backing up data from storage device 20 to storage device 30. Figure 8 is a diagram showing an example of the operational flow of storage device 30 when backing up data from storage device 20 to storage device 30.

At predetermined intervals, the transmission control unit 28 references the differential bitmap stored by the information management unit 25 and checks whether data on the storage medium 100 has been updated. The predetermined time, which is the time interval for checking whether data has been updated, is set appropriately based on, for example, the amount or frequency of data stored from the server 10 to the storage device 20. If no updated blocks have been found (No in step S21), the transmission control unit 28 waits until the predetermined time has elapsed again.

If data has been updated (Yes in step S21), the determination unit 27 references the map information and determines whether the blocks at the backup destination are contiguous (step S22). After determining whether the blocks at the backup destination are contiguous, the determination unit 27 checks whether there are any other blocks where data has been updated. If there are any other blocks where data has been updated (Yes in step S23), the determination unit 27 performs the process of step S22 on the blocks where data has been updated.

If there are no other updated blocks (No in step S23), the transmission control unit 28 starts the process of transmitting the backup data.

If the determination unit 27 determines that the backup destination blocks are consecutive (Yes in step S24), the transmission control unit 28 controls the data transmission unit 23 and the storage medium 100 to instruct the data transmission unit 23 to write to the storage medium and to transmit the backup data to the storage device 30 (step S25).

If the determination unit 27 determines that the backup destination blocks are not consecutive (No in step S24), the transmission control unit 28 controls the data transmission unit 23 and the storage medium 100 to instruct the data transmission unit 23 to write to the extended cache memory and to transmit the backup data to the storage device 30 (step S26). When the backup destination blocks are not consecutive, that is, when the backup destination blocks in the storage device 30 are far apart.

After sending the backup data, the transmission control unit 28 checks whether any of the updated blocks have not yet been sent. If there are any blocks that have not yet been sent (Yes in step S27), the transmission control unit 28 repeats the operations from step S24. If all of the data in the updated blocks has been sent (No in step S27), the storage device 20 waits until it is time to perform the next backup process. The next backup process may be performed, for example, when a preset time has elapsed since the previous backup. The storage device 20 may also perform backup processing when it receives a backup instruction from the server 10.

The write control unit 36 of the storage device 30 receives the instruction for the data write destination and the backup data from the storage device 20 via the data receiving unit 33 (step S31). Upon receiving the instruction for the data write destination, the write control unit 36 checks whether the instruction for the data write destination is the extended cache memory 120 or the storage medium 110.

If the instruction to write the data to a storage medium is Yes in step S32, the write control unit 36 controls the storage medium 110 so that the received backup data is written to the specified block, and writes the backup data received from the storage device 20 to the storage medium 110 (step S38). After writing the backup data to the storage medium 110 is complete, the write control unit 36 checks whether there is any backup data that has not been written. If there is any backup data that has not been written (Yes in step S39), the write control unit 36 repeats the operations from step S32. If there is no backup data that has not been written (No in step S39), the write control unit 36 waits until the next backup instruction is sent.

If the instruction to write the data to is the extended cache memory 120 (No in step S32), the write control unit 36 controls the extended cache memory 120 to write the received backup data to the extended cache memory 120 (step S33).

The write control unit 36 checks the operating status of the storage medium 110 and checks the frequency of access to the storage medium. If the access frequency is higher than the standard (No in step S35), the write control unit 36 continues to monitor the access frequency until the access frequency falls below the standard.

If the access frequency is below the standard (Yes in step S35), the write control unit 36 controls the extended cache memory 120 and storage medium 110 to write the data in the extended cache memory 120 to the storage medium 110 (step S36).

When the backup data in the extended cache memory 120 has been written to the storage medium 110, the write control unit 36 erases the backup data that has been written to the storage medium 110 from the extended cache memory 120. After erasing the backup data, the write control unit 36 checks whether the extended cache memory 120 contains backup data that has not yet been written to the storage medium 110. If there is backup data that has not yet been written (Yes in step S37), the write control unit 36 repeats the operations from step S35. If there is no data that has not yet been written (No in step S37), the write control unit 36 waits until the next backup instruction is sent.

The backup source storage device 20 acquires mapping information for the backup destination storage device 30 and selects either the extended cache memory 120 or the storage medium 110 as the data write destination depending on whether the backup destination blocks are contiguous. However, suppose the mapping status of the backup destination is unknown, and all data is written to the backup destination cache memory, and then written from the cache memory to the storage medium. In such a case, if the amount of data to be backed up increases, the cache memory of the destination storage device may run out of free space during processing. When the cache memory runs out of free space, data must be frequently written from the cache memory to the storage medium to ensure free space in the cache memory. This places a high load on the destination storage medium and may result in a decrease in processing speed. Furthermore, if the cache memory continues to be full, delays may occur in the transmission of backup data from the backup source storage device to the backup destination storage device, and the backup time specifications set for the system may not be met. In particular, if the backup destination storage medium is slow, a high load state is likely to persist, potentially extending the backup time.

In this type of configuration, the backup source storage device 20 of the information processing system of this embodiment obtains mapping information from the backup destination storage device 30 using the thin provisioning function, and determines whether the backup destination blocks are distant from consecutive blocks. If the write destination blocks are consecutive, the backup source storage device 20 sends an instruction to the backup destination storage device 30 to write directly to the storage medium, thereby reducing the load on the storage medium caused by writing data from the extended cache memory to the storage medium. In this way, by using the mapping information obtained from the backup destination to determine whether the backup destination blocks are consecutive and selecting the data write destination to reduce the load on the storage medium, the information processing system of this embodiment can prevent a decrease in backup processing speed.

Second Embodiment
A second embodiment of the present invention will be described in detail with reference to the drawings. Fig. 9 is a diagram showing an outline of the configuration of a storage device 200 of this embodiment. The storage device 200 of this embodiment includes an information management unit 201, a data management unit 202, a determination unit 203, and a transmission control unit 204.

The information management unit 201 acquires mapping information from the backup destination storage device. The data management unit 202 manages, on a block-by-block basis, whether data in the storage area has been updated since the previous backup. The determination unit 203 determines, based on the mapping information, whether the backup destination for the updated block data is a contiguous block in the backup destination storage device. If the backup destination is a contiguous block, the transmission control unit 204 controls the transmission of an instruction to write to the backup destination storage medium and the backup data to the backup destination storage device. If the backup destination is a distant block, the transmission control unit 204 controls the transmission of an instruction to write to the extended cache memory and the backup data to the backup destination storage device.

The operation of the storage device 200 of this embodiment will now be described. Figure 10 is a diagram showing an example of the operation flow of the storage device 200.

The information management unit 201 acquires mapping information from the backup destination storage device (step S201). The data management unit 202 manages whether data in the storage area has been updated since backup on a block-by-block basis (step S202). The determination unit 203 determines, based on the mapping information, whether the backup destination for the updated block data is a contiguous block in the backup destination storage device (step S203). If the backup destination is a contiguous block (Yes in step S204), the transmission control unit 204 controls the transmission of an instruction to write to the backup destination storage medium and the backup data to the backup destination storage device (step S205). If the backup destination is a distant block (No in step S204), the transmission control unit 204 controls the transmission of an instruction to write to the extended cache memory and the backup data to the backup destination storage device (step S206).

In this embodiment, the storage device 200 obtains mapping information from the backup destination storage device and determines whether the backup destination blocks are separate from consecutive blocks. When the write destination blocks are consecutive, writing directly to the storage medium reduces the load on the storage medium caused by writing data from cache memory to the storage medium, thereby preventing a decrease in backup processing speed.

REFERENCE SIGNS LIST 10 Server 20 Storage device 21 Control unit 22 Acquisition unit 23 Data transmission unit 24 Data management unit 25 Information management unit 26 Information storage unit 27 Determination unit 28 Transmission control unit 30 Storage device 31 Control unit 32 Output unit 33 Data reception unit 34 Area management unit 35 Mapping information storage unit 36 Write control unit 100 Storage medium 110 Storage medium 120 Extended cache memory 200 Storage device 201 Information management unit 202 Data management unit 203 Determination unit 204 Transmission control unit

Claims (10)

an information management means for managing mapping information acquired from the backup destination storage device;
a data management means for managing whether or not data has been updated since the previous backup in units of blocks in the storage area;
a determining means for determining whether the backup destination of the updated block data is a continuous block in the backup destination storage device based on the mapping information;
a transmission control means for controlling the transmission of an instruction to write to a storage medium of the backup destination and the backup data to the storage device of the backup destination when the backup destination is a continuous block, and for controlling the transmission of an instruction to write to an extended cache memory and the backup data to the storage device of the backup destination when the backup destination is a distant block. the information management means manages each block of the storage area and a backup destination block in the backup destination storage device by associating them with each other based on the mapping information;
The storage device according to claim 1 . the data management means checks whether the data has been updated since the previous backup when a preset time has elapsed since the previous backup.
3. The storage device according to claim 1. the data management means manages whether or not each block has been updated since the previous backup as a differential bitmap;
4. The storage device according to claim 1. an output means for transmitting mapping information of the storage area to the storage device that is the source of the backup data;
a data receiving means for receiving, from the source storage device, backup data and information indicating whether the backup data is to be written to an extended cache memory or a storage medium;
a write control means for controlling, when the information indicates that the backup data is to be written to the extended cache memory, to hold the received backup data in the extended cache memory and then write the received backup data from the extended cache memory to the storage medium, and, when the information indicates that the backup data is to be written to the storage medium, to write the received backup data to the storage medium. the write control means controls the backup data stored in the extended cache memory to be written to the storage medium when the access frequency to the storage medium is equal to or lower than a predetermined standard, and controls the backup data stored in the extended cache memory to be erased after writing to the storage medium has been completed.
The storage device according to claim 5 . a first storage device comprising the storage device according to any one of claims 1 to 4;
a second storage device comprising the storage device according to claim 5 or 6;
the information management means of the first storage device manages the mapping information acquired from the second storage device;
the transmission control means of the first storage device transmits a write destination instruction and the backup data to the second storage device;
the write control means of the second storage device stores the backup data based on the instruction of the write destination.
Storage system. Manages mapping information acquired from the backup destination storage device,
The system manages whether data has been updated since the last backup in the storage area on a block-by-block basis,
determining whether the backup destination of the updated block data is a continuous block in the backup destination storage device based on the mapping information;
If the blocks are consecutive, control is performed so that an instruction to write to the backup destination block and the backup data are sent to the backup destination storage device, and if the blocks are distant, control is performed so that an instruction to write to the extended cache memory and the backup data are sent to the backup destination storage device.
Backup method. Each block of the storage area is managed in association with a backup destination block in the backup destination storage device based on the mapping information.
The backup method according to claim 8. Sending mapping information of the storage area to the storage device that is the source of the backup data;
receiving, from the source storage device, backup data and information indicating whether the backup data is to be written to an extended cache memory or a storage medium;
When the information indicates that the backup data is to be written to the extended cache memory, the received backup data is stored in the extended cache memory and then written from the extended cache memory to the storage medium, and when the information indicates that the backup data is to be written to the storage medium, the received backup data is written to the storage medium.
Backup method.