JP4877006B2 - Disk array device and data backup system - Google Patents

Description

  The present invention relates to a disk array device and a data backup system, and more particularly to a technique for providing a backup / restore method using a logical disk partial duplication function by a disk array.

  Recently, as the capacity of a logical disk that can be recognized by an operating system increases, the time required for backup and restoration tends to increase in the backup operation of a disk array having a logical disk duplication function in the apparatus.

  Even when multiple logical configurations (disk partitions, applications, database tablespaces, etc.) are placed on a single logical disk for operation, backup and restore operations cannot be performed in logical units. Since it takes a lot of time to backup and restore the entire disk, there is a high possibility that it will affect business operations. A method for solving such problems is required.

  On the other hand, as a conventional technique related to backup and restoration of the entire logical disk of the disk array device, for example, the logical disk described in Patent Document 1 (see particularly paragraphs 0008-0009 etc.) is divided into a plurality of areas, respectively. There was prior art to copy.

  Further, for example, the disk array device described in Patent Document 2 (particularly, refer to paragraphs 0020-0022, etc.) is divided by a common capacity, and each divided area is defined as a slice and copied to this slice. There are conventional techniques for performing various operations such as the above.

However, in both of the prior arts described in Patent Documents 1 and 2, the entire logical disk is the target of backup / restore.
JP 2001-175422 A JP 2006-024024 A

  Backup (meaning copying from the source logical disk to the destination logical disk) and restoration (meaning copying from the destination logical disk to the source logical disk) of the disk array that has the logical disk duplication function in the device In the past, only the entire logical disk could be replicated in operation, so multiple logical configurations built in the source logical disk (disk partitions, database table areas, etc. are referred to as partitions hereinafter) ) Existed, there were the following problems.

  First, since data is copied in units of logical disks during restoration, all partitions other than the partition to be restored may be restored and overwritten with old data.

  Secondly, in restoration, it is necessary to stop the operations related to all partitions on the replication source logical disk because of the necessity of ensuring the consistency of the entire logical disk.

  Thirdly, since it is necessary to maintain consistency as a whole logical disk in the backup, it is necessary to stop all operations related to all partitions on the replication source logical disk.

  Fourthly, with the increase in the capacity of the logical disk, it took an enormous amount of time for backup and restoration.

  In view of the above-described problems of the prior art, an object of the present invention is to provide a backup and restore function for a part of a logical disk in a disk array having a logical disk replication function.

In order to achieve the above object, the invention according to claim 1 is based on a plurality of logical disks including a plurality of partitions, a partition management table for recording partition information of each logical disk , and a logical information based on the partition information. And a partial replica creating means for replicating a partition included in the disk to a partition of another logical disk .

  According to a second aspect of the present invention, in the disk array device according to the first aspect, the partition management table has a flag indicating one of backup processing completion and incomplete for each partition of each logical disk. When the flag indicates that backup processing has been completed for a certain partition, the partial replica creation means determines that restoration processing from the replication destination logical disk to the replication source logical disk is valid for the partition. The partition is replicated from a disk to the replication source logical disk.

Plurality invention of claim 3, including a de Isukuarei device, a server operating the disk array device, but a data communicably connected data backup system, said disk array device, a plurality of partitions A logical disk, a partition management table for recording partition information of each logical disk, and a partial replication creating means for replicating a partition included in a certain logical disk to a partition of another logical disk based on the partition information; And the server has logical disk duplication operation means for instructing the partial duplication creation means to duplicate the partition.

  According to a fourth aspect of the present invention, in the data backup system according to the third aspect, the partition management table has a flag indicating one of backup processing completion and incomplete for each partition of each logical disk. When the flag indicates that backup processing has been completed for a certain partition, the partial replica creation means determines that restoration processing from the replication destination logical disk to the replication source logical disk is valid for the partition. The partition is replicated from a disk to the replication source logical disk.

  According to a fifth aspect of the present invention, in the data backup system according to the third or fourth aspect, the disk array device includes partition information management means for managing the partition management table, and the server includes the partition information management. Instructing the means, the partition information of each logical disk of the disk array device is recorded in the partition management table.

  According to the present invention, a backup and restore function of a part of a logical disk is provided in a disk array having a logical disk replication function.

  Hereinafter, a configuration of a preferred embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an example of the overall configuration of the present embodiment.

  Referring to FIG. 1, the data backup system of this embodiment is connected to a server 1 and a server 1 on which business applications and the like operate, and to the server 1 so as to be able to perform data communication via a fiber channel. And a disk array 2 having a function).

  First, the configuration of the server 1 will be described. The server 1 includes a partition information operation unit 11, a pair setting operation unit 12, and a logical disk duplication operation unit 13 that operate on the server 1, and each unit operates as follows.

  The partition information operation means 11 reads the management information of the replication source logical disk recognized by the operating system, analyzes the partition information, and obtains the logical disk number, partition number, partition start address, and end address information of the replication source logical disk. It has a function of instructing the partition information management means 21 of the disk array to record it in the partition management table 24 in the disk array.

  In addition, the partition information management unit 21 is instructed to return information, and the returned information (replication source logical disk number, partition number, partition start, end address, and replication destination logical disk number (if a pair is configured)) Has a display function.

  The pair setting operation unit 12 has a function of instructing the pair information management unit 22 of the disk array to record pair information of the replication source logical disk and the replication destination logical disk in the partition management table 24 in the disk array.

  The logical disk duplication operation means 13 has a function of instructing the partial replica creation means 23 of the disk array to perform backup from the duplication source logical disk to the duplication destination logical disk and restore from the duplication destination logical disk to the duplication source logic.

  When a partition number is designated together with a replication source logical disk and a replication destination logical disk as an operation target, the partial replication creation means 23 in the disk array is instructed to perform backup and restoration in units of partitions.

  In addition to various business applications running on the server 1, the server 1 also runs an operating system (not shown) for causing the server 1 to function as a server. The copy source logical disk 25 of the array 2 is recognized (“copy source logical disk recognized by the operating system” 14).

  The server 1 realizes various functions for operating the disk array 2 such as the partition information operation means 11, the pair setting operation means 12, and the logical disk duplication operation means 13 in addition to various business applications and operating systems. The server program is running. In the present embodiment, backup of each partition is possible by linking this server program and the disk array 2. In addition, if this server program is a program different from the operating system, it is advantageous from the viewpoint of maintenance such as updating. However, the server program is not limited to this, and the operating system may have some or all functions. Good.

  Next, the configuration of the disk array 2 will be described. The disk array 2 includes a partition information management unit 21, a pair information management unit 22, a partial replication creation unit 23, a partition management table 24, a replication source logical disk 25, and a replication destination logical disk 26. To work.

  The partition information management unit 21 has a function of managing information on the logical disk number, partition number, partition start address, and end address of the replication source logical disk designated by the partition information operation unit 11 in the partition management table 24.

  The pair information management unit 22 has a function of managing information on the pair of logical disks designated by the pair setting operation unit 12.

  The partial copy creation means 23 has a function of performing logical disk backup and restore processing in response to the operation instructions (backup and restore) specified by the logical disk copy operation means 13.

  In addition, when the partition number is specified together with the replication source logical disk and the replication destination logical disk as the operation target, the partial replication creation means 23 starts and ends the specified partition or partitions in the logical disk. Only the data within the address range is copied. If the partition number is not specified, all data on the logical disk is copied.

  For example, in the case where the partition management table 24 has the configuration illustrated in FIG. 2, when the partition number n is specified together with the replication source logical disk number YYYY and the replication destination logical disk number ZZZZZ, backup and restore processing for the partition n is performed. Done.

  In particular, in the case of backup processing, a backup completion flag in the partition management table 24 is set when copying is completed, and it is recorded that the partition to which the flag is set can be restored.

  The disk array 2 includes a replication source logical disk 25 and a replication destination logical disk 26 as physical secondary storage devices of the “replication source logical disk recognized by the operating system” 14 of the server 1 described above. Yes. By this duplication source logical disk 25 and duplication destination logical disk 26, the “duplication source logical disk recognized by the operating system” 14 is duplexed by a method such as mirroring or duplexing.

  Next, with reference to FIGS. 3 to 7, the operation procedure of the present embodiment, in particular, the procedure for operation preparation, backup, and restore operations will be described.

  First, the operation preparation operation of this embodiment will be described. FIG. 3 is a procedure for preparing for operation according to the present embodiment. In the operation preparation operation of this embodiment, partition information is stored in the disk array before the operation is started.

  Specifically, the operating system recognizes the copy source logical disk 25 in step S301 in FIG. Next, in step S302, the partition information management means 21 creates a partition. This operation is realized by, for example, an operation in which the partition information management unit 21 rewrites an entry in the partition management table 24. Next, in step S303, the partition information operation unit 11 executes the partition management unit 21 to store the partition information of the replication source logical disk 25 in the disk array 2. Next, in step S304, the disk array 2 creates a logical disk having the same capacity as the copy source logical disk 25 (this is the copy destination logical disk 26). Next, in step S305, a pair setting procedure, which will be described later with reference to FIGS. 9 and 12, is executed to set the pair of the replication source logical disk 25 and the replication destination logical disk 26 to the disk array 2 (substantially partition management). Store in table 24).

  Next, the backup operation of this embodiment will be described. Note that the present embodiment provides a first backup operation for performing backup of the entire logical disk and a second backup operation for performing backup for each partition in the logical disk. FIG. 4 is a flowchart showing the procedure of the first backup operation. FIG. 5 is a flowchart showing the procedure of the second backup operation.

  In the first backup operation, first, in step S401 of FIG. 4, the disk array 2 is instructed to copy from the replication source logical disk 25 to the replication destination logical disk 26 by executing a replicate command. Next, in step S <b> 402, the server 1 stops all business applications that are using the copy source logical disk 25. In step S403, the disk array 2 is instructed to be separated from the replication source logical disk 25 and the replication destination logical disk 26 by executing a separate command. Next, in step S <b> 404, the server 1 resumes the business application that is using the copy source logical disk 25.

  On the other hand, in the second backup operation, first, in step S501 of FIG. 5, the replication command is executed to instruct the disk array 2 to copy the designated partition from the replication source logical disk 25 to the replication destination logical disk 26. In step S502, the server 1 stops only the business application that uses the backup target partition. In step S503, the disk array 2 is instructed to be separated from the replication source logical disk 25 and the replication destination logical disk 26 by executing a separate command. Next, in step S504, the server 1 resumes the business application that uses the backup target partition.

  In the first backup procedure, it is necessary to stop all the operations using the copy source logical disk in step S402 in FIG. 4, but in the second backup procedure, as shown in step S502 in FIG. Stop only the tasks that use the partition to be backed up.

  Next, the restore operation of this embodiment will be described. Note that this embodiment provides a first restore operation that restores the entire logical disk and a second restore operation that restores each partition in the logical disk. FIG. 6 is a flowchart showing the procedure of the first restore operation. FIG. 7 is a flowchart showing the procedure of the second restore operation.

  In the first restore operation, first, in step S601 of FIG. 6, the server 1 stops all business applications that are using the copy source logical disk 25. Next, in step S602, execution of the restore command instructs the disk array 2 to copy from the replication destination logical disk 26 to the replication source logical disk 25. Next, in step S603, all business applications using the copy source logical disk 25 are restarted.

  On the other hand, in the second restore operation, first, in step S701 in FIG. 7, the server 1 stops only the business application that uses the partition to be restored. Next, in step S702, execution of the restore command instructs the disk array 2 to copy the designated partition from the replication destination logical disk 26 to the replication source logical disk 25. In step S703, the business application that uses the restoration target partition is resumed.

  In FIG. 6, in the first restore procedure, a partition that is not a recovery target is also a copy target and may be overwritten with old data. However, in FIG. 7, in the second restore procedure, there is no risk of copying only the recovery target partition and overwriting the non-recovery target partition with old data.

  Further, in the first restore procedure, it is necessary to stop all the operations using the logical disk to be restored in step S601 in FIG. 6, but in the second restore procedure, in step S701 in FIG. As shown in, it is only necessary to stop the work that uses the partition to be backed up.

  Even in the first restoration procedure, restoration is performed by designating a logical disk, but since the validity of the partition data is recorded in the disk array, only the valid partition portion is copied.

  Next, the operation of each unit shown in FIG. 1 will be described in more detail with reference to FIGS.

  First, the operation of the partition information operation means 11 will be described. FIG. 8 is a flowchart showing the operation of the partition information operation means 11. In FIG. 8, the partition information operation means 11 reads management information existing in the copy source logical disk recognized by the operating system periodically or by command execution, and analyzes the partition information.

  Specifically, first, in step S801, the partition information operating unit 11 checks whether or not the input command or the like is the setting of partition information. If the partition information is set, next, in step S802, the partition information of the logical disk is read. Next, in step S803, partition information is analyzed. Next, in step S804, it is checked whether or not there are a plurality of partitions. If they do not exist, the operation is terminated. If it exists, in step S805, the partition information management unit 21 is instructed to hold the partition information.

  If the partition information is not set in step S801, it is checked in step S806 whether the partition information is displayed. If not, the operation is terminated. If the display is partition information, in step S807, the partition information management unit 21 is instructed to return the partition information. In step S808, it is checked whether there are a plurality of partitions and there is still partition information to be displayed. If there is no partition information, the operation is terminated. If there is, partition information is displayed in step S809.

  Next, the operation of the pair setting operation means 12 will be described. FIG. 9 is a flowchart showing the operation of the pair setting operation means 12. In FIG. 9, the pair setting operation means 12 instructs the pair information management means 22 to record a pair of logical disks (step S901).

  Next, the operation of the logical disk duplication operation means 13 will be described. FIG. 10 is a flowchart showing the operation of the logical disk duplication operation means 13. In FIG. 10, the logical disk duplication operation means 13 gives a backup instruction to the partial duplication creation means 23 when backup is instructed, and gives a restore instruction to the partial duplication creation means 23 when restore is instructed. Do.

  Specifically, first, in step S1001, it is checked whether or not the command is a replicate instruction. If it is a replication instruction, it is checked in step S1002 if a partition is specified. If it is designated, next, in step S1003, the partial replication creating means 23 is instructed to replicate in units of partitions. If not specified, next, in step S1004, the partial replication creating means 23 is instructed to replicate in units of logical disks.

  If it is determined in step S1001 that the command is not a replicate instruction, then in step S1005, it is checked whether or not the command is a restore instruction. If the command is not a restore instruction, the operation ends. In the case of a restore instruction, next, in step S1006, it is checked whether or not a partition is designated. If it is designated, in step S1007, the partial replica creation means 23 is instructed to restore in units of partitions. If not specified, next, in step S1008, the partial replica creation means 23 is instructed to restore in units of logical disks.

  The operation of the logical disk duplication operation means 13 is summarized as follows. When a partition number is designated together with a duplication source logical disk and a duplication destination logical disk as an operation target, the partial duplication creation means 23 is instructed to perform backup and restore in units of partitions. To do.

  Next, the operation of the partition information management unit 21 will be described. FIG. 11 is a flowchart showing the operation of the partition information management means 21. In FIG. 11, the partition information management means 21 secures the partition management table according to the request from the partition information operation means 11 (step S1101), and at the same time, designates the specified replication source logical disk number, partition number, partition start address, partition The end address is recorded in the partition management table 24 (step S1102).

  Next, the operation of the pair information management unit 22 will be described. FIG. 12 is a flowchart showing the operation of the pair information management unit 22. In FIG. 12, the pair information management unit 22 searches the replication source logical disk information (step S1201), and if the search is successful (step S1202, Yes), records the replication destination logical disk number in the partition management table 24 (step S1202). Step S1203).

  Next, the operation of the partial copy creation means 23 will be described. FIG. 13 is a flowchart showing the operation of the partial copy creation means 23. In FIG. 13, when the partition number is specified together with the copy source logical disk 25 and the copy destination logical disk 26 as the operation target, the partial copy creation means 23 designates one or more specified partitions in the logical disk. Only the data within the range of the start and end addresses is copied, and if the partition number is not specified, all the data on the logical disk is copied. For backup processing, set a backup completion flag that is managed in units of partitions when copying is completed, and record that restoration is possible for the partition with the flag set. For restore processing, this flag is ON and replication is enabled. After inspection, copy processing is performed.

  Specifically, first, in step S1301, the partial copy creation unit 23 checks whether the command is a replicate instruction. In the case of a replication instruction, it is checked in step S1302 if a partition number is specified and a command is issued. If issued, then in step S1303, only the designated partition and management information of the copy source logical disk 25 are copied to the copy destination logical disk 26. In step S1304, the backup completion flag for the designated partition number in the partition management table 24 is turned ON.

  If the command is issued without specifying the partition number (No in step S1302), the entire copy source logical disk 25 is copied to the copy destination logical disk 26 in step S1305. In step S1306, the backup completion flags for all partition numbers in the partition management table 24 are turned ON.

  If the command is not a replicate instruction (No in step S1301), then in step S1307, it is checked whether the command is a restore instruction. If the command is not a restore instruction, the operation ends. In the case of the restore instruction, next, in step S1308, it is checked whether or not a command is issued together with the designation of the partition. If there is no designation (step S1308, No), then in step S1312, the entire copy destination logical disk 26 is checked. Is copied to the replication source logical disk 25 and the operation is terminated.

  If a restore instruction is issued together with the designation of the partition (step S1308, Yes), next, in step S1309, it is checked whether or not the partition information exists in the partition management table 24. Exit. If it exists (step S1309, Yes), it is checked in step S1310 if the backup completion flag in the partition management table 24 is ON. If it is not ON (step S1310, No), it cannot be restored because it has not been backed up, and the operation ends. If it is ON (step S1310, Yes), then in step S1311, only the designated partition and management information of the replication destination logical disk 26 are copied to the replication source logical disk 25.

  Next, the effect of this embodiment will be described. According to the present embodiment, the following effects are obtained when a plurality of logical configurations (partitions) exist in a logical disk in backup and restore operations of a disk array having a logical disk replication function in the apparatus.

  First, since it is possible to restore some partitions on the logical disk, it is possible to prevent the partition data that is not subject to restoration from becoming outdated.

  Secondly, since backup is performed in units of partitions, it is not necessary to stop a business using an unrelated partition, so that a backup operation can be flexibly performed according to the business stop time.

  Third, a part of the partitions on the logical disk can be backed up and restored, and the operation stop time can be minimized.

It is a block diagram which shows the structural example of embodiment by this invention. It is a figure shown with the function relevant to the example of a data structure of the partition management table of embodiment by this invention. It is a flowchart which shows the procedure of the operation preparation operation | movement of embodiment by this invention. It is a flowchart which shows the procedure of the 1st backup operation | movement of embodiment by this invention. It is a flowchart which shows the procedure of the 2nd backup operation | movement of embodiment by this invention. It is a flowchart which shows the procedure of the 1st restore operation | movement of embodiment by this invention. It is a flowchart which shows the procedure of the 2nd restore operation | movement of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the partition information operation means of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the pair setting operation means of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the logical disk duplication operation means of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the partition information management means of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the pair information management means of embodiment by this invention. It is a flowchart which shows the procedure of the operation | movement by the partial replication production | generation means of embodiment by this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Server 11 Partition information operation means 12 Pair setting operation means 13 Logical disk replication operation means 14 Replication source logical disk recognized by operating system 2 Disk array 21 Partition information management means 22 Pair information management means 23 Partial replication creation means 24 Partition management Table 25 Copy source logical disk 26 Copy destination logical disk

Claims (5)

Multiple logical disks containing multiple partitions ;
A partition management table for recording partition information of each logical disk;
A disk array device, comprising: a partial replication creating means for replicating a partition included in a certain logical disk to a partition of another logical disk based on the partition information. The partition management table has a flag indicating either backup process completion or incomplete for each partition of each logical disk,
When the flag indicates that the backup process is complete for a certain partition, the partial replica creation means determines that the restore process from the copy destination logical disk to the copy source logical disk is valid for the partition. 2. The disk array device according to claim 1, wherein the partition is copied from the logical disk to the replication source logical disk. And de Isukuarei device, a server operating the disk array device, but a data communicably connected data backup system,
The disk array device
Multiple logical disks containing multiple partitions;
A partition management table for recording partition information of each logical disk;
Based on the partition information, a partial replication creating means for replicating a partition included in a certain logical disk to a partition of another logical disk ,
The server
A data backup system comprising logical disk duplication operation means for instructing the partial duplication creation means to duplicate a partition. The partition management table has a flag indicating either backup process completion or incomplete for each partition of each logical disk,
When the flag indicates that the backup process is complete for a certain partition, the partial replica creation means determines that the restore process from the copy destination logical disk to the copy source logical disk is valid for the partition. The data backup system according to claim 3, wherein the partition is copied from the logical disk to the copy source logical disk. The disk array device
Partition information management means for managing the partition management table;
The server
5. The data backup system according to claim 3, wherein the partition information management means is instructed to record partition information of each logical disk of the disk array device in the partition management table.

Images