JP7103671B2 - Information processing equipment - Google Patents

Description

The present invention relates to an information processing device .

Patent Document 1 discloses a technique in which a system restart can be easily executed by setting a virtual similar disk based on a disk image of a snapshot of a file system at a certain point in time.

Japanese Unexamined Patent Publication No. 11-134117

When updating a volume, there is known a technique for backing up the volume by recording the difference data of the volume due to the update. In order to read the volume, it is necessary to refer to the recorded difference data and the volume that does not correspond to the difference data.
However, since access to the volume occurs in the above reference, the processing time of the reference may be delayed.
An object of the present invention is to provide an information processing apparatus that solves the above-mentioned problems.

The information processing apparatus according to the present invention has a recording means for recording the difference data between before and after updating the data of the first volume in the second volume, and the frequency of occurrence of the update in the first volume based on the difference data. A specific means for specifying a high-frequency update area, which is a storage area higher than a predetermined frequency threshold, and a second recording means for recording clone data of data related to the specified high-frequency update area in the second volume. A generation means for generating a snapshot of the first volume using either the data of the first volume or the clone data is provided , and the specific means compares a plurality of the snapshots with each other. Identify the high frequency update area.

According to at least one of the above aspects, when the difference data of the volume is recorded and backed up, the processing time associated with updating and reading the volume can be shortened.

It is a schematic block diagram which shows the structure of the information processing system which concerns on one Embodiment. It is a figure which shows an example of the tree management table which concerns on one Embodiment. It is a figure which shows an example of the difference management table which concerns on one Embodiment. It is a figure which shows an example of the conversion table which concerns on one Embodiment. It is a figure which shows the concept of the snapshot and the difference which concerns on one Embodiment. It is a figure which shows the concept of the snapshot and the difference which concerns on one Embodiment. It is a figure which shows the concept of the advance copy which concerns on one Embodiment. It is a flowchart which shows the operation of the read command in the snapshot control part which concerns on one Embodiment. It is a flowchart which shows the operation of the read command in the difference control part which concerns on one Embodiment. It is a flowchart which shows the update operation in the snapshot control part when the target volume of the update command which concerns on one Embodiment is a master volume. It is a flowchart which shows the update operation in the difference control part when the target volume of the update command which concerns on one Embodiment is a 2nd volume. It is a flowchart which shows the operation of the reference of the data string stored in the 1st volume, the snapshot volume and the 2nd volume which concerns on one Embodiment. It is a flowchart which shows the operation of the update of the data string stored in the master volume and the 2nd volume which concerns on one Embodiment. It is a flowchart which shows the operation of storing the data string with respect to the 1st volume, the snapshot volume and the 2nd volume which concerns on one Embodiment. It is a flowchart which shows the operation of the advance copy which concerns on one Embodiment. It is a schematic block diagram which shows the structure of the information processing system which concerns on one Embodiment. It is a flowchart which shows the operation of the information processing system which concerns on one Embodiment. It is a schematic block diagram which shows the structure of the computer which concerns on at least one Embodiment.

<First Embodiment>
<< Configuration of information processing system >>
Hereinafter, the information processing system 1 according to the first embodiment will be described in detail with reference to the drawings.
FIG. 1 is a schematic block diagram showing a configuration of the information processing system 1 according to the first embodiment.
The information processing system 1 receives a command from the host computer 20, records the difference data of the volume, and performs backup.
The information processing system 1 includes an information processing device 10, a host computer 20, and an interface 30.

The host computer 20 receives an input from the user of the information processing system 1 and transmits a command to the information processing device 10.
The host computer 20 connects to the information processing device 10 via the interface 30. The host computer 20 may be wirelessly connected to the information processing device 10.

<< Configuration of information processing equipment >>
Hereinafter, the configuration of the information processing device 10 will be described.
The information processing device 10 includes a transfer control unit 11, a snapshot control unit 12, a difference control unit 13, a copy control unit 14, an update control unit 15, and a storage unit 16.

The storage unit 16 is a storage medium that stores a data string in units of fixed-length physical blocks. Hereinafter, the block of the storage unit 16 is referred to as a physical block.
Examples of the storage unit 16 include a magnetic storage device and a semiconductor storage device. The storage unit 16 is not limited to a single physical storage medium, and may be, for example, a virtual storage medium using RAID (Redundant Array of Independent Disks) technology or the like.

A first volume (not shown), a snapshot volume (not shown), and a second volume (not shown), which are virtual storage media, are provided in the area of the storage unit 16. The snapshot volume is further provided with virtual storage media such as snapshot 301 and snapshot 302. The second volume is further provided with a virtual storage medium for storing difference data, clone data, and the like.

The first volume, the snapshot volume, and the second volume store data strings in units of fixed-length logical blocks. The blocks of the first volume, the snapshot volume, and the second volume are referred to as logical blocks.

The logical block may have a different block length from the physical block.
The address indicating the storage destination of the block is referred to as LBA (Logical Block Address). The storage destination of the logical block is referred to as a logical LBA. The storage destination of the physical block is referred to as a physical LBA.
When reading and updating a logical block, the logical LBA is used to specify the target logical block. Further, when reading or updating a physical block, the physical LBA is used to specify the target physical block.
Examples of updates include writing and deleting.

The information processing device 10 includes a memory (not shown). The memory stores the tree management table 900, the difference management table 1000, and the conversion table 1100.

The transfer control unit 11 connects the interface 30, the snapshot control unit 12, and the difference control unit 13. The snapshot control unit 12 connects the difference control unit 13, the copy control unit 14, the update control unit 15, and the storage unit 16.

The transfer control unit 11 receives from the host computer 20 a group of commands including a snapshot acquisition command, a difference acquisition command, an update command, and a read command.

The snapshot control unit 12 receives a snapshot acquisition command from the transfer control unit 11 and generates a snapshot of the first volume. The snapshot control unit 12 is an example of the generation means.

The difference control unit 13 receives a command for acquiring the difference from the transfer control unit 11 and records the difference data due to the update of the first volume in the second volume. The difference control unit 13 is an example of the first recording means.

Further, the snapshot control unit 12 and the difference control unit 13 receive and process a group of commands including an update command and a read command. The update control unit 15 receives and processes a group of commands including a group of commands including a storage unit update command and a storage unit read command from the snapshot control unit 12 and the difference control unit 13.

The difference acquisition command includes the target volume and returns the execution result. The update command includes the target volume, the target logic LBA, and the data string, and returns the execution result. The read command includes the target volume and the target logical LBA and returns the data string. The command for updating the storage unit includes the target physical LBA and the data string, and returns the execution result. The command to read the storage unit includes the target physical LBA and returns the data string.

If the received command is a snapshot acquisition command, the transfer control unit 11 transmits the received command to the snapshot control unit 12 and returns the returned execution result. If the received command is a difference acquisition command, the transfer control unit 11 transmits the received command to the difference control unit 13 and returns the returned execution result. Further, when the received command is an update command, the transfer control unit 11 transmits to the snapshot control unit 12 if the target volume is the master volume, and returns the returned execution result. Further, when the received command is an update command, the transfer control unit 11 transmits to the difference control unit 13 if the target volume is the second volume, and returns the returned execution result. Further, when the received command is a read command, the transfer control unit 11 transmits to the snapshot control unit 12 if the target volume is the master volume, and returns the returned data string.
Further, when the received command is a read command, the transfer control unit 11 transmits to the difference control unit 13 if the target volume is an entity volume or a second volume, and returns the returned data string.

When the received command is a storage unit update command, the update control unit 15 records the data string in the physical block corresponding to the target physical LBA and returns the execution result. If the received command is a command to read the storage unit, the update control unit 15 reads the data string from the physical block corresponding to the target physical LBA and returns the read data string.

FIG. 2 is a diagram showing an example of the tree management table 900. Further, FIG. 3 is a diagram showing an example of the difference management table 1000. Further, FIG. 4 is a diagram showing an example of the conversion table 1100. 2, FIG. 3, and FIG. 4 will be described in detail later.
When the received command is a snapshot acquisition command, the snapshot control unit 12 performs the following operations.

When the received command is a snapshot acquisition command, the snapshot control unit 12 adds new snapshot information to the tree management table 900, sets the target volume as the parent volume, the new snapshot as the child volume, and the type. Store a "snapshot". Further, the snapshot control unit 12 adds a new snapshot entry to the difference management table 1000, stores "0" indicating no difference in all the bits on the entry, and enters the new snapshot in the conversion table 1100. Is added, and "null" indicating no memory is stored in all physical LBAs on the entry, and then the execution result is returned.

If the received command is a difference acquisition command, the difference control unit 13 adds new difference data information to the tree management table 900, sets the target volume in the parent volume, the new difference in the child volume, and "difference" in the type. Store. In addition, a new difference data entry is added to the difference management table 1000, "0" indicating no difference data is stored in all the bits on the entry, a new difference data entry is added to the conversion table 1100, and the entry is on. After storing "null" indicating no memory in all physical LBAs of, the execution result is returned. The processing of the read command and the update command in the snapshot control unit 12 and the difference control unit 13 will be described in detail later.

The copy control unit 14 identifies the high-frequency update area by comparing the bitmaps corresponding to the snapshots for each of a plurality of generations. The copy control unit 14 is an example of the specific means. The high-frequency update area is a storage area in the first volume in which the update occurrence frequency is higher than a predetermined frequency threshold value.
Further, the copy control unit 14 records the clone data of the data related to the specified high-frequency update area in the second volume. The copy control unit 14 is an example of the second recording means.

The copy control unit 14 calculates the update frequency for the difference data registered in the tree management table 900, and if the update frequency exceeds a predetermined threshold value, the LBA or an LBA in the vicinity including the LBA is included. The data string of is recorded from the master volume to the second volume. This series of processing is called pre-copying. The pre-copy process performed by the copy control unit 14 will be described in detail later.

For example, the copy control unit 14 may periodically operate a predetermined program to perform pre-copying at a specified time interval or at a specified time. When the transfer control unit 11 receives some command from the host computer 20, the transfer control unit 11 issues a command to the copy control unit 14, and the copy control unit 14 receives the command and makes a pre-copy. You may try to do.

Next, a detailed configuration of the tree management table 900 shown in FIG. 2, the difference management table 1000 shown in FIG. 3, and the conversion table 1100 shown in FIG. 4 will be described.

The tree management table 900 manages information on a parent volume, a child volume, and a type for all pairs having a parent-child relationship. The parent-child relationship indicates a relationship in which the master volume is the parent and the snapshot volume is the child in the snapshot, and the entity volume is the parent and the second volume is the child in the difference data. For each set, the parent volume is the volume that is the parent of the parent-child relationship, the child volume is the volume that is the child of the parent-child relationship, and the type is, for example, "snapshot" if the parent-child relationship is a snapshot. , If it is difference data, "difference" is stored respectively.
In addition, the old and new relationships are identified according to the order of registration in the tree management table 900 between snapshots acquired from the same master volume or between difference data acquired from the same entity volume. For example, in the snapshot registered in the young number and the snapshot registered in the old number, the old number side is new.

The difference management table 1000 has an entry (a portion surrounded by a broken line in FIG. 3) for each snapshot and difference, and manages whether or not there is a difference with respect to a logical block corresponding to a certain logical LBA. Each entry indicates whether or not there is a difference for all logical LBAs of the corresponding volume with 1 bit. For example, if there is a difference in the data string of the logical block corresponding to a certain logical LBA, "1". , And if there is no difference, store "0".

The conversion table 1100 has an entry (the part surrounded by the broken line in FIG. 4) for each master volume, snapshot volume, and second volume, and corresponds to any physical LBA of the data string of the logical block corresponding to a certain logical LBA. Manage whether the physical block to be stored is stored. Each entry stores the physical LBA that stores the data string for all logical LBAs of the corresponding volume. For example, if the data string of the logical block corresponding to a certain logical LBA does not store the physical block corresponding to any physical LBA, "null" is stored, and if the physical block of the physical LBA 0x0000 stores the data string. Stores "0x0000".

5 and 6 are diagrams showing the concept of snapshots and differences.
Specifically, the duplicate tree is created by acquiring snapshots 301 to 303 with the first volume having four logical blocks as the master volume and further acquiring the difference 401 and the difference 402 with the snapshot 302 as the entity volume. It is configured.
In these figures, the "data" corresponding to each logical LBA on each volume represents a data string presented to the host computer 20 by the logical block corresponding to each logical LBA on the volume.
Further, the data string enclosed in parentheses indicates that the data string is presented as a result of retroactive reference to the newer snapshot volume or master volume because there is no difference in the snapshot volume or the second volume.
The "difference" corresponding to each logical LBA on each volume represents a bit corresponding to each logical LBA in the entry corresponding to the volume in the difference management table 1000.

In the first embodiment, a case where a copy-on-write method is adopted as the snapshot method will be described. That is, the snapshot control unit 12 does not duplicate the data string at the time of taking the snapshot, and when the data string is updated for the master volume, if there is no difference in the latest snapshot volume, the master volume After duplicating the above old data string as a difference on the snapshot volume, the data string on the master volume is updated by the write data string.

On the other hand, when the snapshot control unit 12 reads the data string from the snapshot volume, it starts from the target snapshot, identifies the presence or absence of the difference in the order of the oldest snapshot to the newest snapshot, and the difference exists first. Refer to the snapshot you made. If there is no difference in any of the snapshot volumes, refer to the master volume. The present invention does not limit the snapshot method to the copy-on-write method. For example, the data string on the master volume, which is called the redirect-on-write method, is not updated and the master volume is not updated. A method may be used in which the above data string is duplicated on the snapshot volume and then the data string on the snapshot volume is updated by the write data string.
That is, the snapshot control unit 12 reproduces the data related to the high-frequency update area based on the clone data, and the data related to the storage area other than the high-frequency update area is the data of the first volume or the snapshot volume and the difference data. It is reproduced based on the above, and processing is performed for the read command or the update command. The snapshot control unit 12 is an example of the reproduction means.

The difference control unit 13 does not duplicate the data string when the difference data is acquired, and when the data string is written to the second volume, if there is no difference in the second volume, the missing data on the entity volume After duplicating the column as difference data on the second volume, the data string on the second volume is updated by the write data string.
On the other hand, when reading the data string from the second volume, the difference control unit 13 identifies the presence / absence of the difference data, and if the difference data exists, the second volume is used, and if the difference data does not exist, the entity volume is used. Refer to each.

FIG. 7 is a diagram showing the concept of pre-copying.
The copy control unit 14 executes pre-copy for some reason and records the clone data. A snapshot S from which a certain difference data is created is specified, and from a snapshot (Sm) m generation older than the snapshot S to a snapshot (S + n) n generations newer than the snapshot S (Sm). ), (Sm + 1), ..., (S-1), S, (S + 1), ..., (S + n-1), (S + n) to all logical LBAs on the second volume. On the other hand, the following processing is executed (m and n are natural numbers).
Further, the copy control unit 14 is a number in which the bit of the entry on the difference management table 1000 corresponding to the logical LBA on the snapshots (Sm) to S to (S + n) is "1" for a certain logical LBA. Calculate F. When F exceeds a predetermined threshold value T, the data string of the logical block corresponding to the logical LBA is recorded as clone data in the second volume from the master volume or snapshot volume as the retroactive reference destination ( T is a real number greater than or equal to 0).

In the pre-copy, m, n, and T can take arbitrary values. For example, m, n, and T may be fixedly specified as part of some program in the copy control unit 14. Alternatively, when the transfer control unit 11 receives a command from the host computer 20 including at least one of m, n, and T, the transfer control unit 11 issues a command to the copy control unit 14, and the copy control unit 14 corresponds to the command. When a command is received, m, n, and T included in the command are stored in some area, and the stored m, n, and T can be referred to and used in the pre-copy. m, n, and T are not limited to the configurations according to the above description, and may have different configurations.

<< Operation of information processing system >>
The operation of the information processing system 1 will be described below.
8 to 14 are flowcharts showing the operations of the snapshot control unit 12 and the difference control unit 13 of the information processing system 1.

FIG. 8 is a flowchart showing the operation of the read command in the snapshot control unit 12.
In reading by the snapshot control unit 12, the type of the target volume is identified. Specifically, all the entries on the tree management table 900 are referred to, and the entry in which the target volume is stored as a parent volume or a child volume is searched. If there is an entry in which the target volume is stored as the parent volume and the type of the entry is "snapshot", the target volume is the master volume. If there is an entry in which the target volume is stored as the parent volume and the type of the entry is "difference", the target volume is an entity volume. If there is an entry in which the target volume is stored as a child volume and the type of the entry is "snapshot", the target volume is a snapshot volume. If there is an entry in which the target volume is stored as a child volume and the type of the entry is "difference", the target volume is the second volume. In this way, the snapshot control unit 12 identifies the type of the target volume.

If the target volume is not a snapshot volume (S1001: NO), the process proceeds to S1005.
On the other hand, when the target volume is a snapshot volume (S1001: YES), the target volume is set in the search generation, the oldest snapshot is searched to the latest generation in order from the new snapshot, and the logical LBA is the same as the target logical LBA. The presence or absence of difference data in the snapshot volume in the corresponding logical block is identified (S1003, S1004, S1009).

Specifically, all the entries on the tree management table 900 are referred to, the entry in which the target volume is stored as a child volume is searched, and the parent volume of the entry is identified. Subsequently, all the entries including the entry and existing in the older number than the entry are referred to, and the entries in which the parent volume is stored as the parent volume are sequentially searched from the youngest number to the oldest number, and the searched entries are searched. By sequentially identifying the child volumes of the target generation as snapshots of the target generation, it is possible to search up to the latest generation. Also, referring to the entry on the difference management table 1000 corresponding to the reference logic LBA on the snapshot volume of the target generation, if the bit is "1", the difference exists, and if the bit is "0", the difference is. It is identified as non-existent (hereinafter, the same applies when identifying the presence or absence of a volume difference).

The search ends when a difference exists (S1003: NO), the target logic LBA in the snapshot in which the difference first exists is specified as the reference logic LBA, and the data string is referred to and used as the read data string (S1008, See FIG. 12 for details).

On the other hand, when there is no difference data in any of the snapshot volumes (S1004: YES), the type of the first volume is identified (S1005, S1010).
When the master volume is a snapshot volume (S1005: YES), a read command in which the target volume included in the read command is changed to the master volume is sent to the snapshot control unit 12, and the returned data string is read. (S1006).

When the master volume is the second volume (S1005: NO, S1010: YES), the read command in which the target volume included in the read command is changed to the second volume is transmitted to the difference control unit 13 and returned. The data string is read and used as the data string (S1011).
When the master volume is neither the snapshot volume nor the second volume (S1010: NO), the target logical LBA in the master volume is specified as the reference logical LBA, and the data string is referred to as the read data string (S1012, See FIG. 12 for details).
Finally, the read data string is returned (S1007).

FIG. 9 is a flowchart showing the operation of the read command in the difference control unit 13.
If the target volume is not the second volume (S1101: NO), the process proceeds to S1103.
On the other hand, when the target volume is the second volume, the presence / absence of the difference data of the second volume in the logical block corresponding to the same logical LBA as the target logical LBA is identified (S1102).

When the difference data exists (S1102: NO), the target logic LBA in the second volume is designated as the reference logic LBA, and the data string is referred to as the read data string (S1106, see FIG. 12 for details).
On the other hand, when the difference data does not exist, the type of the entity volume is identified (S1103, S1107).

When the entity volume is a snapshot volume (S1103: YES), a read command in which the target volume included in the read command is changed to the master volume is sent to the snapshot control unit 12, and the returned data string is read data. It is a row (S1104).

When the entity volume is the second volume (S1103: NO, S1107: YES), the read command in which the target volume included in the read command is changed to the second volume is transmitted to the difference control unit 13 and returned. The data string is read and used as the data string (S1108).

When the entity volume is neither the snapshot volume nor the second volume (S1107: NO), the target logical LBA in the entity volume is specified as the reference logical LBA, and the data string is referred to as the read data string (S1109, details). See FIG. 12).
Finally, the read data string is returned (S1105).

FIG. 10 is a flowchart showing the update operation in the snapshot control unit 12 when the target volume of the update command is the master volume. The update made here is a write.

In writing in the snapshot control unit 12, the master volume is specified as the copy source volume, the target logical LBA is specified as the copy source logical LBA, the latest snapshot is specified as the copy destination volume, and the target logical LBA is specified as the copy destination logical LBA. Then, the data string of the logical block corresponding to the same logical LBA as the target logical LBA is duplicated (S1201, see FIG. 13 for details).

Further, the data string is stored with the master volume as the storage volume, the target logic LBA as the storage logic LBA, and the write data string as the storage data string (S1202, see FIG. 14 for details). Finally, the execution result is returned (S1203).

FIG. 11 is a flowchart showing the update operation in the difference control unit 13 when the target volume of the update command is the second volume. The update made here is a write.

In the writing in the difference control unit 13, the entity volume is designated as the replication source volume, the target logic LBA is designated as the replication source logic LBA, the second volume is designated as the replication destination volume, and the target logic LBA is designated as the replication destination logic LBA. The data string of the logical block corresponding to the same logical LBA as the target logical LBA is duplicated (S1301, see FIG. 13 for details).

Further, the data string is stored with the second volume as the storage volume, the target logic LBA as the storage logic LBA, and the write data string as the storage data string (S1302, see FIG. 14 for details). Finally, the execution result is returned (S1303).

FIG. 12 is a flowchart showing the operation of referencing the data strings stored in the first volume, the snapshot volume, and the second volume.
In the data column reference, the entry on the conversion table 1100 corresponding to the reference logical LBA on the reference volume is referred to, and the physical LBA storing the data string of the logical block corresponding to the reference logical LBA is identified (S1401). ..

The identified physical LBA is designated as the target physical LBA, a command for reading the storage unit is transmitted to the update control unit 15, and the returned data string is used as the read data string (S1402).
Finally, the read data string is returned (S1403).

FIG. 13 is a flowchart showing the operation of updating the data strings stored in the master volume and the second volume.
In this case, the information processing apparatus 10 identifies the presence / absence of the snapshot or the difference data of the second volume in the replication destination logic LBA.

If there is difference data in the copy destination volume (S1501: YES), the process proceeds to S1502.
On the other hand, when there is no difference data in the copy destination volume (S1501: NO), the type of the copy source volume is identified (S1503, S1108).
When the copy source volume is a snapshot volume (S1503: YES), the copy source volume is sent to the target volume, the copy source logic LBA is sent to the target logic LBA, and the specified read command is sent to the snapshot control unit 12. The returned data string is used as the replication source data string (S1504).

When the copy source volume is the second volume (S1503: NO, S1508: YES), the copy source volume is set as the target volume, the copy source logic LBA is set as the target logic LBA, and the specified read command is assigned to the difference control unit 13. The data string sent to and returned to the data string is used as the replication source data string (S1509).

If the replication source volume is neither the snapshot volume nor the second volume (S1508: NO), specify the replication source volume as the reference volume and the replication source logical LBA as the reference logical LBA, and refer to the data column. It is used as the copy source data string (S1510, see FIG. 12 for details).
Refer to all the entries in the conversion table 1100, identify the physical LBA that does not store the data string of the logical block corresponding to any logical LBA, and store the specified physical LBA in the entry corresponding to the replication destination logical LBA. do.
That is, the physical block corresponding to the replication destination logical block is acquired (S1505). Further, the data string is stored with the copy destination volume as the storage volume, the copy destination logic LBA as the storage logic LBA, and the copy source data string as the storage data string (S1506, see FIG. 14 for details).

Then, the entry on the difference management table 1000 corresponding to the copy destination logic LBA on the copy destination volume is referred to, and "1" is stored in the bit. That is, it is assumed that there is a difference in the copy destination logic block (S1507). Finally, the execution result is returned (S1502).

FIG. 14 is a flowchart showing the operation of storing the data string for the first volume, the snapshot volume, and the second volume.
The physical LBA is identified by referring to the entry on the conversion table 1100 corresponding to the storage logic LBA on the storage volume (S1601).
The identified physical LBA is designated as the target physical LBA, the stored data string is designated as the data string, the storage unit update command is transmitted to the update control unit 15, and the execution result is received (S1602). Finally, the execution result is returned (S1603).

FIG. 15 is a flowchart showing the operation of pre-copying.
The process related to the pre-copy is executed by designating the target second volume, the target second volume is set as the pre-copy destination second volume, and the entity volume is the creation source of the pre-copy destination second volume. With the original snapshot volume as the target generation, 0 is set in the pre-copy logic LBA (S1701).

If the pre-copy logic LBA is the largest logic LBA on the target second volume (S1702: YES), the execution result is returned (S1717).
On the other hand, if the pre-copy logical LBA is not the maximum logical LBA, set the update frequency F to 0 (S1703), start from the target generation (S1704), and start from a new snapshot created from the same master volume. The oldest snapshot or m generation oldest snapshot is searched in this order, and the number of difference bits of the snapshot in the logical block corresponding to the same logical LBA as the precopy logical LBA is added to the update frequency F. (S1705, S1706, S1707, S1708).

Then, with the target generation as the search generation (S1709), the latest generation or n generation new snapshots are searched in order from the oldest snapshot to the newest snapshot created from the same master volume, and the same as the pre-copy logic LBA. The number of difference bits of the snapshot in the logical block corresponding to the logical LBA of is added to the update frequency F (S1710, S1711, S1712, S1713).

After calculating the update frequency F, if the value obtained by subtracting 1 from F exceeds the threshold value T (S1714: YES), the entity volume that is the creation source of the target second volume is set as the replication source volume, and the entity volume is used. The logical block corresponding to the same logical LBA as the target logical LBA corresponds to the copy source logical LBA, the pre-copy destination second volume corresponds to the copy destination volume, and the pre-copy destination second volume corresponds to the same logical LBA as the target logical LBA. The logical block is designated as the replication destination logical LBA, and the data string of the logical block corresponding to the same logical LBA as the target logical LBA is duplicated (S1715, see FIG. 13 for details).
Finally, 1 is added to the pre-copy logic LBA, and the process proceeds to S1702 (S1716).

《Action / Effect》
The information processing apparatus 10 according to the present invention is a first recording means for recording difference data due to an update of the first volume in the second volume, and a storage area of the first volume in which the update occurrence frequency is higher than a predetermined frequency threshold. A specific means for identifying a certain high-frequency update area, a second recording means for recording the cloned data of the data related to the specified high-frequency update area on the second volume, and the data related to the high-frequency update area based on the clone data. It is provided with a reproduction means for reproducing the data related to the storage area other than the high frequency update area based on the data of the first volume and the difference data.

When updating or reading the backed up data by recording the difference data, the information processing apparatus 10 performs the update or reading using the data related to the area having a high update frequency. Therefore, the user of the information processing apparatus 10 can shorten the processing time associated with updating and reading the volume.

Further, the information processing apparatus 10 includes a generation means for generating a snapshot of the first volume, and the specific means compares a plurality of snapshots to identify a high-frequency update area.

As a result, when the information processing device 10 records the difference data, generates a snapshot, and updates or reads the backed up data, the information processing device 10 uses the data related to the frequently updated area. Update or read. Therefore, the user of the information processing apparatus 10 can shorten the processing time associated with updating and reading the volume.

Further, the identification means of the information processing apparatus 10 identifies the high-frequency update area by comparing the bitmaps corresponding to the snapshots for each of a plurality of generations.
As a result, when the information processing device 10 records the difference data, generates a snapshot, and updates or reads the backed up data, the information processing device 10 relates to an area having a high update frequency by using a bit map. The update or reading is performed using the data. Therefore, the user of the information processing apparatus 10 can shorten the processing time associated with updating and reading the volume.

The information processing method according to the present invention includes a step of recording the difference data due to the update of the first volume in the second volume, and a high frequency update in which the update occurrence frequency of the first volume is higher than a predetermined frequency threshold. The step of specifying the area, the step of recording the clone data of the data related to the specified high-frequency update area in the second volume, and the step of reproducing the data related to the high-frequency update area based on the clone data, and the high-frequency update area. It has a step of reproducing the data related to the storage areas other than the above based on the data of the first volume and the difference data.

When updating or reading the backed up data by recording the difference data, if the information processing method is used, the update or reading can be performed using the data related to the area having a high update frequency. Therefore, the user who uses the information processing method can shorten the processing time associated with updating and reading the volume.

The program according to the present invention includes a step of recording the difference data due to the update of the first volume in the second volume, and a high-frequency update area in the first volume, which is a storage area in which the update occurrence frequency is higher than a predetermined frequency threshold. The step of specifying, the step of recording the clone data of the data related to the specified high-frequency update area in the second volume, and the step of reproducing the data related to the high-frequency update area based on the clone data, other than the high-frequency update area. The data related to the storage area is executed as a step of reproducing based on the data of the first volume and the difference data.

When updating or reading the backed up data by recording the difference data, when the program is executed, the update or reading can be performed using the data related to the area having a high update frequency. Therefore, the user who executes the program can shorten the processing time associated with updating and reading the volume.

<Second embodiment>
<< Configuration of information processing system >>
Hereinafter, the configuration of the information processing system 1 according to the second embodiment will be described.
The copy control unit 14 of the information processing system 1 according to the first embodiment identifies a high-frequency update area by comparing bitmaps corresponding to snapshots for each of a plurality of generations. On the other hand, the copy control unit 14 of the information processing system 1 according to the second embodiment identifies a high-frequency update area by comparing a plurality of difference data.

FIG. 16 is a diagram showing a configuration of the information processing system 1 according to the second embodiment.
The storage unit 16 according to the second embodiment is provided outside the information processing device 10. The storage unit 16 is connected to the information processing device 10 via the interface 30B. Examples of the storage unit 16 according to the second embodiment include a disk array and cloud storage.

Further, since the command for updating the storage unit and the command for reading the storage unit in the second embodiment are transmitted to the storage unit 16 via the transfer control unit 11, compatibility with the external storage unit 16 is required. do.
For example, when the storage unit 16 provides a plurality of volumes, it is necessary to specify the target volume as well.
As described above, as commands for compatibility between a plurality of storage devices such as between the information processing device 10 and the storage unit 16, for example, ATA (Advanced Technology Attainment) specifications and SCSI (Small Computer System Interface) specifications are used. , Commands based on the NVM Express specification can be used. The command is not limited to this as long as the information processing device 10 and the storage unit 16 can recognize each other.

<< Operation of information processing system >>
Hereinafter, the operation of the information processing system 1 according to the second embodiment will be described.
FIG. 17 is a flowchart showing the operation of the second embodiment of the present invention.
The target second volume is specified and executed, and 0 is set in the pre-copy logic LBA (S1801).

If the pre-copy logic LBA is the largest logic LBA on the target second volume (S1802: YES), the execution result is returned (S1818).
On the other hand, when the pre-copy logic LBA is not the maximum logic LBA, the presence or absence of a difference in the target second volume in the logic block corresponding to the pre-copy logic LBA is identified (S1803).

If there is a difference (S1803: NO), the process proceeds to S1817.
On the other hand, if there is no difference, the update frequency F is set to 0 (S1804), the target second volume is started (S1805), and the new second volume is the oldest second volume created from the same entity volume. The number of difference bits of the second volume in the logical block corresponding to the same logical LBA as the pre-copy logical LBA is updated by searching up to the first second volume or the second volume created m times before in the order of volumes. Add to F (S1806, S1807, S1808, S1809).

Subsequently, starting from the target second volume (S1810), the second volume created from the old second volume to the new second volume created from the same entity volume is the last second volume or the second volume created n times later. Is searched up to, and the number of difference bits of the second volume in the logical block corresponding to the same logical LBA as the pre-copy logical LBA is added to the update frequency F (S1811, S1812, S1813, S1814).

After calculating the update frequency F, if the value obtained by subtracting 1 from F exceeds the threshold value T (S1815: YES), the entity volume that is the creation source of the target second volume is set as the replication source volume, and the entity volume is used. The logical block corresponding to the same logical LBA as the target logical LBA corresponds to the copy source logical LBA, the pre-copy destination second volume corresponds to the copy destination volume, and the pre-copy destination second volume corresponds to the same logical LBA as the target logical LBA. The logical block is designated as the replication destination logical LBA, and the data string of the logical block corresponding to the same logical LBA as the target logical LBA is duplicated (S1816, see FIG. 13 for details).
Finally, 1 is added to the pre-copy logic LBA, and the process proceeds to S1802 (S1817).

In the information processing device 10 according to the first embodiment, the entity volume that is the creation source of the pre-copy destination second volume needs to be a snapshot, but the information processing device 10 according to the second embodiment has. Even when the entity volume is not a snapshot volume, pre-copying can be realized by calculating the update frequency from a plurality of second volumes created from the same entity volume.

《Action / Effect》
The identification means of the information processing apparatus 10 according to the present invention identifies a high-frequency update region by comparing a plurality of difference data.
The information processing apparatus 10 can shorten the processing time associated with updating and reading the master volume by calculating the update frequency from the difference data of the second volume without comparing the snapshots.

Further, the second volume of the information processing device 10 is provided outside the information processing device 10, and the second volume and the information processing device 10 are connected by an interface 30.
As a result, the second volume is not necessarily built in the information processing device 10, and a storage medium provided by an external storage device such as a disk array device may be used. As a result, the information processing apparatus 10 can shorten the processing time associated with updating and reading the master volume by using various external second volumes.

<Basic configuration>
The configuration of the embodiment according to the basic configuration is the same as the configuration of the first embodiment described above. That is, the information processing device 10 according to the basic configuration includes a transfer control unit 11, a snapshot control unit 12, a difference control unit 13, a copy control unit 14, an update control unit 15, and a storage unit 16.

The information processing device 10 according to the basic configuration is a first recording means for recording the difference data due to the update of the first volume in the second volume, and a storage area in the first volume in which the update occurrence frequency is higher than a predetermined frequency threshold. A specific means for identifying a certain high-frequency update area, a second recording means for recording the cloned data of the data related to the specified high-frequency update area on the second volume, and the data related to the high-frequency update area based on the clone data. It is provided with a reproduction means for reproducing the data related to the storage area other than the high frequency update area based on the data of the first volume and the difference data.

When updating or reading the backed up data by recording the difference data, the information processing apparatus 10 performs the update or reading using the data related to the area having a high update frequency. Therefore, the user of the information processing apparatus 10 can shorten the processing time associated with updating and reading the volume.

<Computer configuration>
FIG. 18 is a schematic block diagram showing the configuration of a computer according to at least one embodiment.
The computer 2100 includes a processor 2110, a main memory 2120, a storage 2130, and an interface 2140.
The information processing device 10 described above is mounted on the computer 2100. The operation of each processing unit described above is stored in the storage 2130 in the form of a program. The processor 2110 reads a program from the storage 2130, expands it into the main memory 2120, and executes the above processing according to the program. Further, the processor 2110 secures a storage area corresponding to each of the above-mentioned storage units in the main memory 2120 according to the program.

The program may be intended to realize some of the functions exerted by the computer 2100. For example, the program may exert its function in combination with another program already stored in the storage 2130, or in combination with another program mounted on another device. In another embodiment, the computer 2100 may include a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device) in addition to or in place of the above configuration. Examples of PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, some or all of the functions realized by the processor 2110 may be realized by the integrated circuit.

Examples of the storage 2130 include magnetic disks, magneto-optical disks, semiconductor memories, and the like. The storage 2130 may be internal media directly connected to the bus of the computer 2100, or external media connected to the computer via interface 2140 or a communication line. When this program is distributed to the computer 2100 via a communication line, the distributed computer 2100 may expand the program in the main memory 2120 and execute the above processing. In at least one embodiment, the storage 2130 is a non-temporary tangible storage medium.

Further, the program may be for realizing a part of the above-mentioned functions. Further, the program may be a so-called difference file (difference program) that realizes the above-mentioned function in combination with another program already stored in the storage 2130.

1 Information processing system 10 Information processing device 11 Transfer control unit 12 Snapshot control unit 13 Difference control unit 14 Copy control unit 15 Update control unit 16 Storage unit 20 Host computer 30 Interface 2100 Computer 2110 Processor 2120 Main memory 2130 Storage 2140 Interface

Claims (3)

A recording means for recording the difference data between before and after updating the data of the first volume in the second volume, and
Based on the difference data, a specific means for identifying a high-frequency update area in the first volume, which is a storage area in which the update occurrence frequency is higher than a predetermined frequency threshold value, and
A second recording means for recording cloned data of the identified data related to the high-frequency update area in the second volume, and
A generation means for generating a snapshot of the first volume using either the data of the first volume or the clone data, and
With
The identifying means identifies the frequently updated region by comparing a plurality of the snapshots.
Information processing device. The information processing apparatus according to claim 1 , wherein the specific means identifies the high-frequency update area by comparing bitmaps corresponding to the snapshots for each of a plurality of generations. The second volume is provided outside the information processing apparatus.
The information processing device according to claim 1 or 2 , wherein the second volume and the information processing device are connected by an interface.

Images