JP5720366B2 - File management system and backup method - Google Patents

Description

  The present invention relates to a file management system and a backup method capable of efficiently performing file backup.

  In the data backup, when the number of data increases, the backup time becomes longer. As a method of shortening the backup time, there is a method of backing up only a file whose contents have been changed since the previous backup (hereinafter referred to as a differential file) without backing up all of the data. Hereinafter, backing up only files whose contents have been changed is referred to as differential backup.

  It is necessary to detect a differential file when performing a differential backup. In order to shorten the detection time of the difference file, virtual file data including summary data and path data of the actual file data is generated, and it is determined whether or not there is a difference in the actual file data based on the virtual file data. Thus, there has been proposed a method that eliminates the need to scan the actual file data itself and can efficiently determine the presence or absence of a difference (see, for example, Patent Document 1). In addition, there is a method of generating a backup file of the target file every time the file is changed (for example, see Patent Document 2).

JP 2008-234269 A JP 2010-231656 A

  In the method described in Patent Document 1, the same number of virtual file data is required with respect to the number of real file data. Therefore, when the number of real file data is large, it takes time to detect the difference. In the method described in Patent Document 2, it is not necessary to detect the difference at the time of backup. However, since the target file is backed up to the backup storage unit every time the file is changed, it takes time to change the file. Cost.

  Therefore, the present invention is characterized by providing a file management system and a backup method capable of reducing the processing time and data area required for differential backup.

A file management system according to the present invention is a file management system for managing a user's file, an update detection means for determining whether or not a file operation by a user terminal is creation or update of a file, In the case of creation or update, link creation means for creating and storing a link having the file to be created or updated as an entity file and backup processing means for storing the file based on the link as an update file in the backup device The link creation means acquires path information of the file to be created or updated from the update detection means, converts the path information into a character string not including directory information, and converts the character string to the file name of the link characterized by a.

A backup method according to the present invention is a backup method in a file management system for managing a user's file, and determines whether or not a file operation by a user terminal is creation or update of a file, and the file operation is a If it is an update, the path information of the file to be created or updated is converted into a character string that does not include directory information, the character string is used as a file name, and the file to be created or updated is referred to as an entity file. A link is created and stored, and a file based on the link is stored in the backup device as an update file.

  According to the present invention, the processing time and data area required for differential backup can be reduced.

It is a block diagram which shows the structure of the file management system by this invention. It is explanatory drawing which shows the structure of the file system part in 1st Embodiment. It is explanatory drawing which shows the structure of the backup directory operation part in 1st Embodiment. It is explanatory drawing which shows the structure of the restore operation part in 1st Embodiment. It is a flowchart which shows an example of the process which accumulate | stores the difference file of the file management system by this invention. It is a flowchart which shows an example of the differential backup process of the file management system by this invention. It is a flowchart which shows an example of the differential restore process of the file management system by this invention. It is a block diagram which shows the principal part of the file management system by this invention.

Embodiment 1. FIG.
A first embodiment of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing the configuration of a file management system according to the present invention.

  As shown in FIG. 1, the file management system includes a file server 200. The file server 200 can access the backup device 300. The user terminal 400 is communicably connected to the file server 200, and accesses the file server 200 by using a function of a CIFS (Common Internet File System) client or an NFS (Network File System) client. Although one user terminal 400 is shown in FIG. 1, generally, many user terminals can access the file server 200.

  The file server 200 includes a file server function unit 210, a file system unit 220, a backup directory operation unit 240, a restore operation unit 250, and a storage unit 260.

  The file server function unit 210 has a file server function such as a CIFS server or an NFS server. The file server function unit 210 receives an access request for the client storage area 261 of the storage unit 260.

  When the client storage area 261 is accessed, the file server function unit 210 receives file access information and path detection information, and passes the access information and path detection information to the file system unit 220. The access information is information indicating a file operation such as update or deletion of a file or directory. The path detection information is information for detecting path information of a file to be accessed.

  Based on the access information received from the file server function unit 210, the file system unit 220 performs processing of reading and writing a file to the storage unit 260. The storage unit 260 includes a backup directory management area 262 in addition to the client storage area 261. The client storage area 261 is an area in which files corresponding to the directory structure disclosed to the user terminal 400 are stored. The backup directory structure provided in the backup directory management area 262 cannot be accessed from the user terminal 400, but the backup device 300 can read the backup directory structure.

  The file system unit 220 includes an update file detection unit 230. The update file detection unit 230 detects a file update operation for the client storage area 261.

  The backup directory operation unit 240 operates the backup directory management area 262.

  The restore operation unit 250 has a function of restoring data stored in the backup device 300.

  The storage unit 260 is a storage medium such as a hard disk.

  The backup device 300 includes a data storage unit 310 such as a backup server or a tape device, for example. The data storage unit 310 stores data stored in the storage unit 260 of the file server 200.

  FIG. 2 is an explanatory diagram illustrating a configuration of the file system unit according to the first embodiment.

  As shown in FIG. 2, the update file detection unit 230 of the file system unit 220 includes an update system operation hook unit 231 and a path information acquisition unit 232.

  The update file detection unit 230 calls the update system operation hook unit 231 when the file system unit 220 receives access information and path detection information from the file server function unit 210.

  The update operation hook unit 231 analyzes the access information. If the access is an update operation for the client storage area 261, the update operation hook unit 231 hooks the process and calls the path information acquisition unit 232.

  The path information acquisition unit 232 generates path information of a file to be accessed from the path detection information and passes it to the backup directory operation unit 240.

  FIG. 3 is an explanatory diagram illustrating the configuration of the backup directory operation unit according to the first embodiment.

  As shown in FIG. 3, the backup directory operation unit 240 includes a directory creation / deletion unit 241, a path information conversion unit 242, a link creation unit 243, and a management information recording unit 244.

  The directory creation / deletion unit 241 creates a work directory in the backup directory management area 262 and creates a management information file 601.

  The path information conversion unit 242 generates path conversion information in which “/” in the path information input to the backup directory operation unit 240 is replaced with another character that does not include the path information. For example, if the path information is “/ABC/DE/FGH.txt”, the path conversion information is “XABCXDEXFGH.txt”, which is a character string that does not include directory information.

  The link creation unit 243 creates a link in the work directory in the backup directory management area 262 using the path conversion information generated by the path information conversion unit 242 as a file name and a file referenced by the path information as an actual file. . A link is a concept that allows one entity file to be referred to by another name or path name. Examples of the links include hard links and symbolic links in UNIX (registered trademark) and Linux (registered trademark).

  The management information recording unit 244 records the access information in the management information file 601 when the access information received by the backup directory operation unit 240 is file deletion, directory creation or deletion.

  FIG. 4 is an explanatory diagram illustrating a configuration of the restore operation unit according to the first embodiment.

  As shown in FIG. 4, the restore operation unit 250 includes a path information restoration unit 251 and a directory restoration unit 252.

  The path information restoration unit 251 converts the file name of the file read from the backup device 300. Specifically, the first character of the file name and the same character as the first character included in the file name are replaced with “/”. By this replacement, the path information is restored, and the file is restored to the restore destination in the client storage area 261 indicated by the path information.

  The directory restoration unit 252 deletes unnecessary directories based on the management information file read from the backup device 300 and creates a necessary directory.

  Next, the operation of this embodiment will be described.

  FIG. 5 is a flowchart showing an example of processing for accumulating a difference file in the file management system according to the present invention.

  As shown in FIG. 5, access to the file server 200 from the user terminal 400 occurs at an arbitrary timing (step S501). The update operation hook unit 231 determines whether the access is a file creation or update process based on the access information of the access in step S501 (step S502).

  If the file access is a file creation or update process (Yes in step S502), the update operation hook unit 231 hooks the process and calls the path information acquisition unit 232. In parallel with other processing of the file system unit 220, the path information acquisition unit 232 performs processing for generating path information of a file to be created or updated. The path information acquisition unit 232 passes the path information to the backup directory operation unit 240. The path information conversion unit 242 of the backup directory operation unit 240 converts the path information to generate path conversion information (step S503).

  The link creation unit 243 uses the path conversion information generated in step S503 as a file name, and creates a link that refers to the file to be created or updated in the work directory of the backup directory management area 262 (step S504). . If a link having the same file name already exists, it is overwritten.

  If the file access is other than file creation or update processing (No in step S502), the update operation hook unit 231 determines whether the file access is file deletion processing (step S505).

  If the file access is a file deletion process (Yes in step S505), the link creation unit 243 confirms whether the link of the file to be deleted exists in the working directory of the backup directory management area 262, and links If exists, the link is deleted (step S506). The management information recording unit 244 records the file deletion processing content in the management information file 601 (step S507). The processing content is the path information of the deleted file. If the file access is other than the file deletion process (No in step S505), the update operation hook unit 231 determines whether the file access is a directory creation or deletion process (step S508).

  When the file access is a directory creation or deletion process (Yes in step S508), the management information recording unit 244 records the directory creation or deletion process contents in the management information file 601 (step S509).

  Note that the processing of the backup directory operation unit 240 in steps S503 to S509 is performed in parallel with the processing of the file system unit 220.

  As described above, a link relating to a file that has been created or updated is created in the backup directory management area 262. That is, when an operation is performed on a file, a link that can specify a file to be differentially backed up is created.

  FIG. 6 is a flowchart showing an example of the differential backup process of the file management system according to the present invention.

  When the backup device 300 performs backup, the file server 200 creates a snapshot of the directory structure and the contents of the backup directory management area 262 that are disclosed to the user terminal 400 in order to secure a quiesce point of the file (step) S601). The backup device 300 backs up data using the work directory in the backup directory management area in the snapshot (step S602). At this time, the contents of the work directory in the backup directory management area 262 are deleted by the file server 200.

  If there is a link in the work directory of the backup directory management area in step S602, it is determined that there is a difference in the link entity file, and the link entity file is acquired from the client storage area and is stored in the backup device 300. Backed up. In the file name of the entity file, the same file name as that of the link, that is, path conversion information is set. After step S602, the file server 200 deletes the snapshot created in step S601.

  Note that when it is possible to prevent access to the storage unit 260 of the user terminal 400 during differential backup execution, it is not necessary to create a snapshot. In this case, the backup is executed using the contents of the backup directory management area 262 as they are.

  Further, when the differential backup is continuously executed, the backup device 300 may manage the area separately for each differential backup.

  FIG. 7 is a flowchart showing an example of differential restore processing of the file management system according to the present invention.

  With reference to FIG. 7, the differential restore processing of the file server 200 will be described. The differential restore is a process for reproducing updates performed during a difference target period (a period from when the previous differential backup is executed until the next differential backup is executed). FIG. 7 shows processing when differential restoration is performed from the state where the client storage area 261 of the file server 200 is restored to the state at the start of the difference target period.

  When the file server 200 executes differential restore, the directory restoration unit 252 reads the management information file from the backup device 300, and deletes unnecessary directories in the restore destination directory according to the recorded contents of the read management information file. Create the necessary directories. If file deletion is recorded in the management information file, the deleted file is deleted from the restore destination directory (step S701).

  The path information restoration unit 251 of the restore operation unit 250 reads the update file from the backup device 300. The update file is a link entity file backed up in step S602. The path information restoration unit 251 converts the file name of the update file, and restores the path information of the update file (step S702).

  The restore operation unit 250 restores (updates or creates) the update file to the client storage area 261 according to the restored path information (step S703). If the same file exists at the restore destination, the file is updated.

  The processes in steps S702 and S703 are executed for all the update files stored in the backup device 300 (step S704).

  As described above, in this embodiment, a differential file is substantially detected by creating a link at the time of file update, and a differential backup is executed by backing up the file for which the link has been created. Therefore, it is not necessary to detect a difference file when executing backup.

  In addition, since the file name of the link is set with a character string that does not include directory information, the links corresponding to the difference files can be consolidated into one directory, and the link is acquired without searching the directory at the time of backup. be able to. In addition, since the link of the file with the difference, not the file with the difference, is stored in the backup directory management area, the processing time can be shortened and the data area required for the differential backup can be reduced. it can.

  FIG. 8 is a block diagram showing the main part of the file management system according to the present invention. As shown in FIG. 8, the file management system according to the present invention is a file management system for managing a user's file, and a file operation by a user terminal (corresponding to the user terminal 400 shown in FIG. 1) is the creation of a file. An update detection means 21 (corresponding to the update file detection unit 230 in the file server 200 shown in FIG. 1) that determines whether or not it is an update, and if the file operation is creation or update of a file, A link creation means 22 (corresponding to the backup directory operation unit 240 in the file server 200 shown in FIG. 1) for creating and storing a link having the file to be updated as an entity file, and the file based on the link as an update file Backup device 30 (corresponding to the backup device 300 shown in FIG. 1) Backup processing unit 23 to be stored (corresponding to the file system unit 220 in the file server 200 shown in FIG.) And a.

  In the above embodiment, the following file management system is also disclosed.

(1) A file management system in which the link creation unit 22 acquires path information of a file to be updated from the update detection unit 21, converts the path information into a character string that does not include directory information, and uses the character string as a path name.

(2) When the file operation is deletion of the file, the link creation unit 22 records the contents of the file operation in the management information file, and the backup processing unit 23 stores the management information file in the backup device 30. File management system.

(3) The link creation means 22 is a file management system that records the contents of the file operation in the management information file when the file operation is creation or deletion of a directory.

(4) The restore operation means (corresponding to the restore operation unit 250 in the file server 200 shown in FIG. 1) acquires the management information file and the update file from the backup device 30, and restores the directory based on the contents of the management information file. And file deletion, restore path information from the file name of the update file, and restore the update file to the restore destination indicated by the path information.

21 Update detection means 22 Link creation means 23 Backup processing means 30, 300 Backup device 200 File server 210 File server function part 220 File system part 230 Update file detection part 231 Update system operation hook part 232 Path information acquisition part 240 Backup directory operation Section 241 Directory creation / deletion section 242 Path information conversion section 243 Link creation section 244 Management information recording section 250 Restore operation section 251 Path information restoration section 252 Directory restoration section 260 Storage section 261 Client storage area 262 Backup directory management area 310 Data Storage unit 400 User terminal 601 Management information file

Claims (5)

A file management system for managing user files,
Update detection means for determining whether the file operation by the user terminal is creation or update of a file;
If the file operation is creation or update of a file, link creation means for creating and storing a link having the file to be created or updated as an entity file; and
Backup processing means for storing a file based on the link in the backup device as an update file ,
The link creation means acquires path information of the file to be created or updated from the update detection means, converts the path information into a character string not including directory information, and converts the character string to the file name of the link file management system which is characterized in that a. If the file operation is a file deletion, the link creation means records the content of the file operation in the management information file,
Backup processing means, the file management system according to claim 1 for storing the management information file to the backup device. Link creation means, the file management system according to claim 1 file operation when was the creation or deletion directory, which records the contents of the file operation in the management information file. Equipped with restore operation means,
The restore operation means acquires a management information file and an update file from a backup device, restores a directory and deletes the file based on the contents of the management information file, and restores path information from the file name of the update file. The file management system according to claim 3 , wherein the update file is restored to the restore destination indicated by the path information. A backup method in a file management system for managing user files,
It is determined whether or not the file operation by the user terminal is a file creation or update. If the file operation is a file creation or update, the path information of the file to be created or updated is obtained as directory information. Converts to a character string that does not include it, creates and stores a link with the character string as a file name and the file to be created or updated as an entity file, and stores the file based on the link as an update file in the backup device A backup method characterized by allowing

Images