JP4389772B2 - Computer system and backup method - Google Patents

Description

  The present invention relates to data backup in a computer system, and more particularly to a backup system that saves update data in another storage every time storage data is updated by an application program.

  In general, the entire storage or the entire file is copied to another storage in preparation for the loss of data accumulated in the storage of the computer system (for example, Patent Documents 1 and 2). However, in such batch saving, when data stored in the storage is lost, it is necessary to return to the copy point. Therefore, every time storage data is updated by an application program, update data is stored in another storage.

  A typical example of a conventional backup method that takes backup every data update is the RAID 1 method of Redundant Arrays of Inexpensive Disks (RAID) (for example, Non-Patent Document 1). This conventional backup method usually uses two disks and writes the same data to both disks, so that if one disk fails, the other can continue processing, improving fault tolerance. It is. Here, a local disk is usually used as the two disks, but a storage area network (SAN) that makes it possible to use a network-connected disk or a disk on a network-connected computer as a local disk and It is also possible to use a disk on the network as one of the disks using a technique such as Network Attached Storage (NAS). By doing so, it is possible to store the replicated data in a remote place, although there is a performance degradation due to communication delay.

Another example of a conventional backup method in which a backup is taken every time data is updated is described in Patent Document 3. In this conventional backup method, every time there is a data update access to the file data from the application program in the master server, the data related to the update location is extracted and transferred to the slave server, and the slave server disk is the same as the master server side. By updating to, the slave server can continue processing even if the master server fails, improving fault tolerance.
Japanese Patent Laid-Open No. 5-313972 JP 2003-316635 A JP 2000-163300 A DA Patterson, G. Gibson, and RH Katz., A case for redundant arrays of inexpensive disks (RAID), Int. Conf. On Management of Data (SIGMOD), pages 109-116, Chicago, IL, USA, June 1988

  The above-mentioned methods have been proposed in the past as a method for taking a backup every time data is updated. However, any of the conventional methods is designed to always save backup data every time storage data is updated by an application program. There is no idea of multiplexing only a part of the stored data. Here, a part of data is a part of a plurality of files stored in the storage, a file updated by a part of application programs, and the like. Of course, it is possible to store the files to be multiplexed and other files in separate storages, and only the storage for storing the files to be multiplexed can be multiplexed by the method of Non-Patent Document 1. Since the processing such as matching is complicated and the number of necessary storages is increased, it is desired to realize a flexible backup method capable of multiplexing only a part of files while being stored in the same storage.

  The present invention has been proposed in view of such circumstances, and the object thereof is to store only update data satisfying a predetermined backup condition in another storage every time storage data is updated by an application program. To provide a backup system.

A first computer system of the present invention is a computer system in which a master computer and one or more slave computers are connected via a network, and the slave computer includes a slave side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the backup storage data transmitted from the computer, and the master computer includes a master-side operating system. A storage for storing data, an application program for updating data in the storage through the master side operating system, and backup condition information including file information Including a backup condition table, a master-side data transmitting / receiving unit for communicating with the slave-side data transmitting / receiving unit, and update operation information for the storage data issued from the application program, which includes file information for specifying an update file Update operation information is acquired through the master side operating system, and the storage is updated through the master side operating system based on the acquired update operation information, and matches the file information included in the acquired update operation information It is determined whether backup condition information including file information to be stored is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that backup is performed, and the acquired update operation information And a backup management unit that based send the same updates and update of the storage updating operation information for performing with respect to the backup storage to said one or more slave computer via the master-side data transceiver.

A second computer system of the present invention is a computer system in which a master computer and one or more slave computers are connected via a network, and the slave computer includes a slave-side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the backup storage data transmitted from the computer, and the master computer includes a master-side operating system. A storage that stores data, an application program that updates data in the storage through the master side operating system, and a backup that includes application program information Backup condition table for storing information, master side data transmitting / receiving unit for communicating with the slave side data transmitting / receiving unit, and update operation information for the storage data issued from the application program, and specifying an update file Update operation information including file information to be acquired is acquired via the master side operating system, the storage is updated through the master side operating system based on the acquired update operation information, and the acquired update operation information is issued It is determined whether backup condition information including application program information that matches the application program information of the application program that has been stored is stored in the backup condition table. If condition information is stored, it is determined that backup is to be performed, and update operation information for performing update similar to the storage update based on the acquired update operation information to the backup storage is provided on the master side A backup management unit that transmits the data to the one or more slave computers via a data transmission / reception unit.

A third computer system of the present invention is a computer system in which a master computer and one or more slave computers are connected via a network, and the slave computer includes a slave side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the backup storage data transmitted from the computer, and the master computer includes a master-side operating system. A storage for storing data, an application program for updating data in the storage through the master side operating system, and a backup condition information including user information. Including a backup condition table, a master-side data transmitting / receiving unit for communicating with the slave-side data transmitting / receiving unit, and update operation information for the storage data issued from the application program, and file information for specifying an update file The update operation information is acquired through the master side operating system, the storage program is updated through the master side operating system based on the acquired update operation information, and the application program that has issued the acquired update operation information is It is determined whether backup condition information including user information that matches the user information of the activated user is stored in the backup condition table. If the corresponding backup condition information is stored, backup is performed. Determine that things, the obtained based on the updated operation information the storage of the update and the same as defined above for an update operation information for performing with respect to the backup storage and update via the master-side data transceiver 1 or more slaves And a backup management unit for transmitting to the computer.

In the first backup method of the present invention, a master computer and one or more slave computers are connected via a network, and the slave computer has been transmitted from a slave side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for data in the backup storage, and the master computer stores the master-side operating system and data A backup condition table for storing storage, an application program for updating data in the storage through the master side operating system, and backup condition information including file information And a master-side data transmission / reception unit for communicating with the slave-side data transmission / reception unit, wherein the backup management unit of the master computer has the storage program issued from the application program. Update operation information with respect to data and including update file information that specifies an update file is acquired via the master side operating system, and the storage is stored through the master side operating system based on the acquired update operation information. In addition to updating, it is determined whether backup condition information including file information matching the file information included in the acquired update operation information is stored in the backup condition table, and the corresponding backup condition information is recorded. If it is determined that backup is to be performed, update operation information for performing update similar to the update of the storage based on the acquired update operation information to the backup storage is sent to the master side data transmitting / receiving unit. To the one or more slave computers.

In the second backup method of the present invention, a master computer and one or more slave computers are connected via a network, and the slave computer has been transmitted from a slave side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for data in the backup storage, and the master computer stores the master-side operating system and data A storage, an application program for updating data in the storage through the master side operating system, and a backup condition information including application program information. A backup method in a computer system comprising an up condition table and a master side data transmitting / receiving unit for communicating with the slave side data transmitting / receiving unit, wherein the backup management unit of the master computer is issued from the application program Update operation information for the data in the storage is acquired through the master-side operating system, and update operation information including file information for specifying an update file is acquired, and through the master-side operating system based on the acquired update operation information Backup condition information including application program information that matches the application program information of the application program that issued the acquired update operation information and updated the storage is stored in the buffer. It is determined whether it is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that the backup is performed, and the same update as the update of the storage based on the acquired update operation information is performed. Is sent to the one or more slave computers via the master side data transmitter / receiver.

In the third backup method of the present invention, a master computer and one or more slave computers are connected via a network, and the slave computer has been transmitted from a slave side operating system, a backup storage, and the master computer. A slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for data in the backup storage, and the master computer stores the master-side operating system and data Backup condition table for storing storage, application program for updating data in the storage through the master side operating system, and backup condition information including user information A backup method in a computer system comprising a master-side data transmission / reception unit for communicating with the slave-side data transmission / reception unit, wherein the backup management unit of the master computer has the storage data issued from the application program The update operation information including the file information for specifying the update file is acquired through the master side operating system, and the storage is updated through the master side operating system based on the acquired update operation information. And whether backup condition information including user information that matches the user information of the user who started the application program that issued the acquired update operation information is stored in the backup condition table. An update for performing the same update as the update of the storage based on the acquired update operation information to the backup storage by determining, determining that the corresponding backup condition information is stored, and performing the backup The operation information is transmitted to the one or more slave computers via the master side data transmitting / receiving unit.

The first computer according to the present invention includes the slave-side operating system, the backup storage, and the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer. A computer connected to one or more slave computers including a slave side data transmitting / receiving unit via a network, a master side operating system, a storage for storing data, and data in the storage as the master Application program to be updated through the operating system, backup condition table storing backup condition information including file information, and master side for communicating with the slave side data transceiver A data transmission / reception unit and update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, through the master side operating system; The storage condition is updated through the master side operating system based on the updated operation information, and backup condition information including file information matching the file information included in the acquired update operation information is stored in the backup condition table. If the relevant backup condition information is stored, it is determined that the backup is performed, and the same update as the storage update based on the obtained update operation information is performed on the backup storage. Do An update operation information of order through the master-side data transceiver and a backup management unit that transmits to the one or more slave computer.

The second computer according to the present invention includes the slave side operating system, the backup storage, and the backup storage via the slave side operating system based on the update operation information for the data of the backup storage transmitted from another computer. A computer connected to one or more slave computers including a slave side data transmitting / receiving unit via a network, a master side operating system, a storage for storing data, and data in the storage as the master An application program to be updated through the operating system, a backup condition table for storing backup condition information including application program information, and communication with the slave data transmitter / receiver A master-side data transmission / reception unit, and update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, is acquired via the master-side operating system And updating the storage through the master-side operating system based on the acquired update operation information and including application program information that matches the application program information of the application program that issued the acquired update operation information It is determined whether information is stored in the backup condition table. If the corresponding backup condition information is stored, it is determined that backup is performed, and the acquired update operation is determined. And a backup management unit that transmits an update operation information for performing the same updates and update of the storage system based on information to said backup storage to said one or more slave computer via the master-side data transceiver.

The third computer of the present invention includes the slave side operating system, the backup storage, and the backup storage via the slave side operating system based on update operation information for the data of the backup storage transmitted from another computer. A computer connected to one or more slave computers including a slave side data transmitting / receiving unit via a network, a master side operating system, a storage for storing data, and data in the storage as the master An application program to be updated through the operating system, a backup condition table for storing backup condition information including application program information, and communication with the slave data transmitter / receiver A master-side data transmission / reception unit, and update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, is acquired via the master-side operating system And updating the storage through the master side operating system based on the acquired update operation information, and including user information that matches the user information of the user who started the application program that issued the acquired update operation information It is determined whether backup condition information is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that backup is performed, and the storage based on the acquired update operation information is determined. And a backup management unit that transmits an update operation information for performing the same updates and di updates to the backup storage to said one or more slave computer via the master-side data transceiver.

"Action"
In the present invention, when an application program operating on a computer issues update operation information for updating data stored in a storage provided in the computer to the operating system, the backup management unit acquires the update operation information. Then, it is determined whether or not any of the backup condition information held in the backup condition table is met, and if it is matched, an update similar to the storage update based on the update operation information is performed on the backup storage. To control. Thus, every time the storage data is updated by the application program, only the update data that satisfies a predetermined backup condition is stored in another storage.

  According to the present invention, every time storage data is updated by an application program, only update data that satisfies a predetermined backup condition can be stored in another storage. The reason is that the backup management unit acquires update operation information issued from the application program to the operating system, determines whether it matches any backup condition information stored in the backup condition table, and This is because control is performed so that the same update as the storage update based on the update operation information is performed on the backup storage.

  Further, according to the present invention, the capacity of the backup storage can be made smaller than that of the backup source storage, and the construction cost of the backup system can be reduced. The reason is that only some of the plurality of files stored in the backup source storage can be used as the backup source.

  Further, according to the present invention, it is not necessary to add any processing to an application program that provides various services by referring to and updating storage. The reason is that the backup management unit that operates independently of the application program acquires update operation information issued to the operating system from the application program, and whether it matches any backup condition information held in the backup condition table. This is because control is performed so that the same update as the storage update based on the update operation information is executed on the backup storage only when it matches.

  Next, embodiments of the present invention will be described in detail with reference to the drawings.

“First Embodiment”
Referring to FIG. 1, the computer system according to the first embodiment of the present invention includes a computer 100 that operates as a master computer and a plurality of backup computers 110-1 to 110-3 that operate as slave computers. 201 are communicably connected to each other. In FIG. 1, there are three backup machines, but at least one backup machine is sufficient.

  The computer 100 includes an application program 101, an operating system (hereinafter referred to as OS) 102, a storage (database) 103, a backup condition table 104, a backup management unit 105, and a data transmission / reception unit 106. Each backup machine 110-1 to 110-3 includes backup storages 111-1 to 111-3, a data transmission / reception unit 112, and an OS 113. Each of these components generally has the following functions.

  An application program 101 operating on the computer 100 refers to and updates data stored in the storage 103, and provides a user with predetermined functions such as a database function, a web server function, and a mail server function. When the application program 101 refers to or updates the storage 103, it issues data operation information to the OS 102. The operation information of the data includes the operation target and the operation content. The operation target is information for specifying a file to be referred to and updated, and the operation content is a type of reference and update. For example, if the OS 102 conforms to UNIX (registered trademark) and POSIX (Portable Operating System Interface for UNIX), the character string str can be written to the file of the file descriptor fd by the command write (fd, str).

  Although only one application program 101 is shown in FIG. 1, there are generally a plurality of application programs. Each application program is activated on the computer 100 by a command line input from a console or the like (not shown), and operates in parallel. Each running application program is called a process and is identified by a process ID. Each process has an execution user (owner) and a group, which are identified by an execution user name and a group name.

  The OS 102 is basic software of the computer 100 and manages and controls computer resources such as execution of the application program 101 and internal storage / external storage.

  The storage 103 is an external storage device configured by a hard disk device or the like, and stores a plurality of files storing data processed by the application program 101. Each file is identified by a file name and stored in a directory that organizes files by purpose or type. Each directory is identified by a directory name. Directories have a tree structure, and paths from the root to individual directories and end files are called paths. A directory name on the path from the root of the tree to a certain file and the file name connected by a predetermined symbol (for example, slash /) is called a path name of the file.

  The backup condition table 104 is a table that holds backup condition information of the storage 103. For example, the backup condition table 104 is held in an external storage device, and is expanded in an internal storage (main memory) when the computer 100 is executed. The backup condition table 104 describes at least one backup condition information. Details of the backup condition information will be described later.

  The backup management unit 105 is update operation information (update operation by write or the like, and open / close operation of a file as a prerequisite) among the operation information for the data in the storage 103 issued from the application program 101 to the OS 102. In addition, operation information matching any backup condition information held in the backup condition table 104 is detected, and backup processing is controlled according to the detection result. As a specific example for realizing such detection and control of backup processing according to the detection, a program code for realizing these processing is included in a system call of the OS 102 used when the application program 101 updates data in the storage 103. There is a method to describe in. In this way, for example, when the function “write” is called in the application program 101, an interrupt is generated and the system call sys_write () is executed inside the kernel of the OS 102, and the backup management unit 105 described above in the system call. Processing is executed. As another implementation example, the program code of the library used when the application program 101 updates the data in the storage 103 is the program code that realizes the processing of the backup management unit 105 although the input / output format is the same. There is a way to replace it with another embedded code. Thus, for example, when the function “write” is called in the application program 101, the above-described processing of the backup management unit 105 embedded in the corresponding code of the library is executed. The backup management unit 105 executes processes such as a process for synchronizing the storage 103 and the backup storages 111-1 to 111-3 and a data recovery process in addition to the above-described detection and the control of the backup process corresponding thereto. To do.

  The data transmission / reception unit 106 receives data update operation information related to the backup process from the backup management unit 105 and transfers it to the backup machines 110-1 to 110-3 through the network 201. The data received from the backup machines 110-1 to 110-3 is transferred to the backup management unit 105.

  The backup storages 111-1 to 111-3 of the backup machines 110-1 to 110-3 are external storage devices configured by a hard disk device or the like, and a copy of data stored in the storage 103 of the computer 100 is stored. Save in units of files. The backup storages 111-1 to 111-3 and the backup storages 111-1 to 111-3 and the storage 103 do not have to be the same storage, but have different capacities and different specifications (hard disk cylinders, tracks, etc. , Sectors, headers, etc.) or having different interface protocols (SCSI, ATA, etc.). This is because backup is performed by multiplexing with higher layer information called update operation information of data issued from the application program.

  The OS 113 is basic software for the backup machines 110-1 to 110-3, and manages and controls computer resources such as execution of application programs (not shown) and internal and external storage.

  The data transmission / reception unit 112 communicates with the data transmission / reception unit 106 of the computer 100 through the network 201. When data update operation information is received from the data transmission / reception unit 106, the update operation information is issued to the OS 113 to update the backup storage 113.

  The network 201 is an arbitrary communication line such as a LAN, a wireless LAN, the Internet, and a public line, and may be a multiplexed communication line that combines a plurality.

  Next, operations performed when the application program 101 updates data in the storage 103 among the operations of the present embodiment will be described in detail with reference to the flowcharts of FIGS. As a premise, it is assumed that a copy of the data in the storage 103 is stored on the backup storage side regarding the data to be multiplexed. This state is that the storage 103 and the backup storages 111-1 to 111-3 are synchronized. A specific method for taking the initial synchronization will be described later.

  First, when data update operation information for the storage 103 is issued from the application program 101 (S1001 in FIG. 2-1), a data update interrupt occurs in the OS 102 or a data update library is called (S1002), and backup management is performed. Control is transferred to the unit 105. The backup management unit 105 analyzes the update operation information of the data issued by the application program 101 and collates with the backup condition information set in the backup condition table 104 (S1003). Then, if there is backup condition information that matches the data update operation information issued by the application program 101 (Yes in S1003), the backup management unit 105 sets the backup storages 111-1 to 111-3 in the same manner as the storage 103. After notifying the data transmission / reception unit 106 of update operation information for updating (S1004), and if there is no suitable backup condition information (No in S1003), the process of step S1004 is skipped and this time through the OS 102 The storage 103 is updated based on the update operation information (S1005), and the data update result is notified to the application program 101 through the OS 102 (S1013, S1014). The application program 101 proceeds with processing according to the update result (S1015).

  On the other hand, the data transmitting / receiving unit 106 transmits the data update operation information passed from the backup management unit 105 to the backup machines 110-1 to 110-3 via the network 201 (S1006). The data transmitting / receiving unit 112 of the backup machines 110-1 to 110-3 updates the backup storages 111-1 to 111-3 through the OS 113 based on the received data update operation information (S1008).

  In this way, only update data that satisfies the backup condition set in the backup condition table 104 is reflected in the backup storage 111-1 to 111-3.

  In the process example of FIG. 2A, the storage update process in step S1005 is performed independently of the processes in steps S1004 to S1008, and therefore the delay of the data update process in the computer 100 can be reduced. However, even if the data in the storage 103 is updated in the computer 100, a state in which the backup storage 111-1 to 111-3 is not updated in the backup machine may occur. There is a possibility that. For this reason, in order to improve reliability, the storage update process in step S1005 may be performed after confirming the completion of the backup storage update process in step S1008. Alternatively, as shown in the flowchart of FIG. 2-2, the backup storage update processing result in step S1008 is notified from the data transmission / reception unit 112 to the data transmission / reception unit 106 (S1009), and further from the data transmission / reception unit 106 to the backup management unit 105. In step S1010, the backup management unit 105 determines whether both the storage update process in step S1005 and the backup storage update process in step S1008 have succeeded (S1011). If it has failed, the processing S1012 for invalidating the storage update on the successful side may be performed. In order to realize the storage update invalidation processing, the data before the update of the storage 103 and the data before the update of the backup storage 111-1 to 111-3 are stored in the internal storage managed by the OS 102 and the OS 113, for example. In a case where the storage update is invalidated, it is possible to write back to the data before the update stored in the internal storage.

  Next, an initial operation for synchronizing the storage 103 and the backup storages 111-1 to 111-3 in this embodiment will be described with reference to the flowcharts of FIGS. 1 and 3.

  The backup management unit 105 indicates that the file stored in the storage 103 and the file serving as the backup of the file stored in the backup storage 111-1 to 111-3 are not synchronized, for example, each file update time If the application program 101 detects that there is a difference in file size (S1101 in FIG. 3), the operation information related to the data update for the backed up file in the storage 103 from the application program 101 is subsequently stored in a buffer (not shown) managed by the OS 102. (S1102). Next, the backup management unit 105 reads the data of the entire area of the file to be backed up in the storage 103 (S1103), and updates operation information including the data of the read area through the data transmitting / receiving unit 106 to the backup machines 110-1 to 111. -3 (S1104). The data transmitting / receiving unit 112 of each backup machine 110-1 to 110-3 receives the update operation information including the above data (S1105), and updates the backup storage 111 via the OS 113 based on the received update operation information. (S1106). The update result is transmitted from the data transmitting / receiving unit 112 of the backup machines 110-1 to 110-3 via the reverse path (S1107) and received by the data transmitting / receiving unit 106 of the computer 110 (S1108). When the size of the read and update area is large, the processing of steps S1103 to S1108 is repeated to synchronize the backup target files of the storage 103 and the backup storage 111.

  If the synchronization process cannot be completed for some reason (No in S1109), an error process such as saving an error log is performed (S1110). If the synchronization processing is completed (Yes in S1109), the update management information for the data that has been newly updated is accumulated in the buffer during the synchronization processing. Reading (S111), the data transmitting / receiving unit 106 transmits the data to the backup machines 110-1 to 110-3 (S1112). When the data transmission / reception unit 112 of the backup machines 110-1 to 110-3 receives the update operation information (S1113), it updates the backup storage 111 via the OS 113 (S1114). The update result is transmitted from the data transmitting / receiving unit 112 of the backup machines 110-1 to 110-3 via the reverse path (S1115) and received by the data transmitting / receiving unit 106 of the computer 110 (S1116). When the processing for the update operation information of all the data held in the buffer is completed, the data update operation information for the backup target file in the storage 103 is subsequently saved in the buffer (S1117), and FIG. Processing is performed by the method described in FIG. Thereby, the storage 103 and the backup storage 111 are updated synchronously.

  Next, a process for restoring a service using a backup storage will be described.

  When the service is resumed using the backup machines 110-1 to 110-3, as shown in FIG. 4, the application program 101 operating on the computer 100 in the backup machine 110 (110-1 to 110-3) is loaded. to start. Since the application data 101 is stored in the backup storage 111 (111-1 to 111-3), the application program 101 uses the data to restart the service.

  On the other hand, in the computer 110, it is also possible to restore data to the storage 103 from the backup storages 111-1 to 111-3 of the backup machines 110-1 to 110-3 and restart the service. Processing in that case will be described with reference to the flowcharts of FIGS.

  When the OS 102 detects a recovery command of the storage 103 input from a console (not shown) (S1201), the control shifts to a recovery process of the backup management unit 105. The backup management unit 105 refers to the backup condition table 104 to determine in which backup machine the latest and necessary data is stored, and notifies the data transmission / reception unit 106 of the backup data transfer request (S1202). ). If the same backup data is in multiple backup machines, you may request transfer to any one backup machine, or transfer to multiple locations and communicate with the backup machine with the fastest response. Alternatively, the transfer request may be made by dividing the backup data into a plurality of locations. The data transmitting / receiving unit 106 transmits a backup data transfer request to the backup devices 110-1 to 110-3 (S1203), and the data transmitting / receiving unit 112 of the backup devices 110-1 to 110-3 receives the request (S1204). The corresponding file data is read from the backup storages 110-1 to 110-3 via the OS 113 (S1205). The read data is transmitted from the data transmitting / receiving unit 112 to the computer 100 (S1206) and received by the data transmitting / receiving unit 106 of the computer 100 (S1207). The backup management unit 105 checks whether or not all data transfer has succeeded (S1208), and if successful, updates the storage 103 using the transferred data (S12-9) and completes the processing. Is returned to the OS 102. On the other hand, when the data transfer has failed (No in S1208), a command result indicating the processing failure is returned to the OS 102.

  The OS 102 receives the result of the command (S1210), and then executes a startup command for the application program 101 (S1211). The application program 101 reads data to be processed from the storage 103, resumes the process (S1212), and returns the process result to the OS (S1213).

  Next, the backup condition table 104 will be described in detail.

  FIG. 6 shows an example of the backup condition table 104. The backup condition table C2000 in this example holds a total of 10 pieces of backup condition information R2001 to R2010. Each piece of backup condition information includes at least information about a file as a backup target condition. Here, the information on the file means a file name, a directory name, a path name, and the like. The example of FIG. 6 uses a backup source path as information about a file as a backup target condition. In the case of FIG. 6, each backup condition information further includes a backup destination machine name, a backup destination path, and a schedule. The backup destination machine name is information for selecting one or several backup machines when there are a plurality of backup machines. The backup destination path is required when backing up to a file with a path different from the backup source path, and is omitted in the same case. The schedule specifies additional backup conditions. Hereinafter, individual backup condition information will be described.

  In the backup condition information R2001, when the path of the file to be updated by the application program 101 matches / home / usr1 set as the backup source path, that is, the application program 101 updates files under the directory of / home / usr1. In this case, it means that backup is performed with the same file path to the backup destination machine name busrv1.

  Unlike the backup condition information R2001, the backup condition information R2002 specifies the IP address “192.168.0.1” as the backup destination machine name. In this case, in addition to the busrv1 backup machine specified by the backup condition information R2001, backup is also performed by the backup machine having the IP address 192.168.0.1. In this way, when a plurality of backup condition information matches, a plurality of backups are created.

  The backup condition information R2003 describes “0 <= $ Hour <12” in the schedule. This means that if the backup execution time is from 0:00 to 12:00, that is, if data update to / home / usr2 is in the morning, the backup destination machine name busrv1 is backed up with the same file path. On the other hand, the backup condition information R2004 is described as “12 <= $ Hour <24” in the schedule, and the backup execution time is from 12:00 to 24:00, that is, data update to / home / usr2 is in the afternoon. Indicates that backup is performed with the same file path to the backup destination machine name busrv2.

  The backup condition information R2005 describes “($ Times mod 3) == 0” in the schedule. This means that if the number of backups is divisible by 3, backup is performed with the same file path to the backup destination machine name busrv1. On the other hand, the backup condition information R2006 is described as “($ Times mod 3) == 1” in the schedule, and when the number of backups is divided by 3 and left by 1, the same file path is used for the backup destination machine name busrv2. When backup is performed and the backup condition information R2007 is described as “($ Times mod 3) == 2” in the schedule and the number of backups is divided by 3 and left by 2, the same file is stored in the backup destination machine name busrv3. This means that backup is performed by path. Note that the cumulative number of times $ Times of data update to each file under the backup source path / home / usr3 is held in a buffer managed by the OS 102, for example.

  In the backup condition information R2008, “/ srv1 / home / usr4” is specified as the backup destination path. This means that the file path is backed up by replacing / home / usr4 at the backup source with / srv1 / home / usr4 at the backup destination. That is, the file /home/usr4/file.txt is backed up with the file name /srv1/home/usr4/file.txt in the backup machine busrv1.

  In the backup condition information R2009, “/ home / usr / [a-m] *” is specified as the backup source path name. Here, [a-m] * is a regular expression that means an arbitrary character string whose first character begins with alphabets a to m. In this backup condition information R2009, when the file is the directory / home / usr and the first character of the file name is an alphabet between a and m, / disk1 / home / usr of the backup machine with the IP address 192.168.0.2 It means to back up. Thus, a regular expression can be specified for the backup source path name. On the other hand, the backup condition information R2010 specifies “/ home / usr / [nz] *” as the backup source path name, the file is the directory / home / usr, and the first character of the file name is between n and z. Means that the backup is to / disk2 / home / usr of the backup machine whose IP address is 192.168.0.2.

  When the backup condition table C2000 as shown in FIG. 6 is used, the backup management unit 105 stores the storage 103 to be updated by the application program 101 in the backup condition matching process S1003 of FIGS. 2-1 and 2-2. The backup condition information having a backup source path that matches the file path is searched from the backup condition table C2000. If there is backup condition information that matches the backup source path, the schedule column is referred to, and if it is set, backup condition information for which the schedule condition is true is extracted. Then, a path name of the backup file is generated from the path of the file to be updated and the backup destination path in the extracted backup condition information, and the backup file with the path name is updated in the same manner as the file of the storage 103. Update operation information is generated and notified to the data transmission / reception unit 106 with information on the backup destination machine name. The data transmitting / receiving unit 106 transmits the notified data update operation information to the backup machine having the notified backup destination machine name.

  FIG. 7 shows another example of the backup condition table 104. The backup condition table C2100 in this example holds a total of three pieces of backup condition information R2101 to R2103. Each piece of backup condition information includes at least information related to the application program as a backup target condition. Here, the information related to the application program means an application program name, a process name, a command line, and the like. In the example of FIG. 7, the application program name is used as information related to the application program as the backup target condition. In the case of FIG. 7, each backup condition information includes a backup destination machine name, a backup destination path, and a schedule, as in the backup condition table of FIG. Hereinafter, individual backup condition information will be described.

  The backup condition information R2101 indicates that when the application program name of the application program 101 that updates the storage 103 is sendmail, a file is backed up under / app / mail of the backup machine busrv1. For example, when sendmail updates the file / var / spool / mail / usr1 in the storage 103, it backs up to / app / mail / var / spool / mail / usr1 of busrv1.

  The backup condition information R2102 indicates that when the application program name of the application program 101 that updates the storage 103 is httpd, the backup of the file is performed under / app / www1 of the backup machine busrv2, and the backup condition information R2103 indicates that the storage 103 When the application program name of the application program 101 that updates the file is db, it indicates that the file backup is performed under / app / db2 of the backup machine busrv3. A regular expression can be used for the application program name.

  When the backup condition table C2100 as shown in FIG. 7 is used, the backup management unit 105 searches for the ID of the data update calling process in the backup condition matching process S1003 of FIGS. 2-1 and 2-2, A process name, that is, an application program name is acquired from the ID. For example, if the OS 102 conforms to UNIX and POSIX, the process ID is obtained using the commands getpid and getppid, and the obtained process ID is obtained from the list information of the running processes obtained using the ps command. This is possible by acquiring the process name corresponding to, that is, the application program name. Then, the backup management unit 105 searches the backup condition information having the acquired application program name from the backup condition table C2100, and if there is matching backup condition information, refer to the schedule column and if it is set, Backup condition information for which the schedule condition is true is extracted. Then, a path name of the backup file is generated from the path of the file to be updated and the backup destination path in the extracted backup condition information, and the backup file with the path name is updated in the same manner as the file of the storage 103. Update operation information is generated and notified to the data transmission / reception unit 106 with information on the backup destination machine name. The data transmitting / receiving unit 106 transmits the notified data update operation information to the backup machine having the notified backup destination machine name.

  FIG. 8 shows still another example of the backup condition table 104. The backup condition table C2200 in this example holds a total of three pieces of backup condition information R2201 to R2203. Each backup condition information includes at least information about the user as a backup target condition. Here, the information regarding the user means a user name, a user ID, a group name, a group ID, and the like. In the example of FIG. 8, a user name and a group name are used as information about a user as a backup target condition. User names and group names can be specified by separating them with dots, and regular expressions can be used for each. In the case of FIG. 8, the individual backup condition information includes a backup destination machine name, a backup destination path, and a schedule, as in the backup condition table of FIG. Hereinafter, individual backup condition information will be described.

  The backup condition information R2201 means that if the user name of the user who started the application program 101 that updates the storage 103 is usr1 and the group name is users, the backup of the file is performed under / usr1 of the backup machine busrv1. For example, when usr1.users updates a file called /home/usr1/file.txt, it backs up to /usr1/home/usr1/file.txt on busrv1.

  The backup condition information R2202 indicates that if the user name of the user who started the application program 101 that updates the storage 103 is mail and the group name is arbitrary, the file is backed up under / mail of the backup machine busrv2. means. Also, the backup condition information R2203 indicates that if the user name of the user who started the application program 101 that updates the storage 103 is arbitrary and the group name is wheel, the file is backed up to / system under the backup machine busrv3. means. Here, * is a regular expression.

  When the backup condition table C2200 as shown in FIG. 8 is used, the backup management unit 105 searches for the ID of the data update calling process in the backup condition matching process S1003 of FIGS. 2-1 and 2-2, The user name and group name are acquired from the ID. For example, if the OS 102 conforms to UNIX and POSIX, the process ID is obtained using the commands getpid and getppid, and the obtained process ID is obtained from the list information of the running processes obtained using the ps command. This is possible by acquiring the user name and group name corresponding to. Then, the backup management unit 105 searches the backup condition table C2200 for backup condition information having the acquired user name and group name, and if there is matching backup condition information, the backup management unit 105 is set by referring to the schedule column. If so, backup condition information for which the schedule condition is true is extracted. Then, a path name of the backup file is generated from the path of the file to be updated and the backup destination path in the extracted backup condition information, and the backup file with the path name is updated in the same manner as the file of the storage 103. Update operation information is generated and notified to the data transmission / reception unit 106 with information on the backup destination machine name. The data transmitting / receiving unit 106 transmits the notified data update operation information to the backup machine having the notified backup destination machine name.

  Next, the effect of this embodiment will be described.

  According to the present embodiment, when the application program 101 updates the data in the storage 103, only the update data that satisfies a preset backup condition is installed in the backup machines 110-1 to 110-3 at remote locations. -1 to 111-3. The reason is that the backup management table 105 is provided with a backup condition table 104 that holds at least one backup condition information. The backup storage information 111-installed in the remote backup devices 110-1 to 110-3 is searched only when the backup condition information that is analyzed and matched is searched from the table 104 and the matching backup condition information is registered. This is because it operates to store a backup of the updated data in 1-111-3. For this reason, the backup storages 111-1 to 111-3 having a smaller capacity than the storage 103 can be used, and the bandwidth of the communication line of the network 201 can be reduced, thereby reducing the construction cost of the remote backup system. Can do.

  If the backup condition table C2000 shown in FIG. 6 is used, backup of some files specified by information about files such as path names, file names, and directory names among a plurality of files stored in the storage 103. Can be stored in remote backup storage. Therefore, if a configuration in which a particularly important file is stored under a certain path among a plurality of files in the storage 103 to be referenced and updated by the application program 101 is adopted, the application program 101 is not modified. Without the application, it is possible to automatically multiplex only those important files.

  If the backup condition table C2100 shown in FIG. 7 is used, it is updated by the application program 101 specified by the information related to the application program such as the application program name and process name among the plurality of files stored in the storage 103. File-only backups can be stored in remote backup storage.

  If the backup condition table C2200 shown in FIG. 8 is used, the application program 101 started by the user specified by the information about the user such as the user name and group name among the plurality of files stored in the storage 103 is used. A backup of only the files to be updated can be stored in a remote backup storage.

“Second Embodiment”
Referring to FIG. 9, in the second embodiment of the present invention, the computer 100 includes a data recording unit 151 instead of the data transmission / reception unit 106, and backup machines 110-1 to 110-3 are connected via the network 201. Instead, the first is that a removable and portable data storage medium 150 is connected to the computer 100 and a backup machine 110 having a backup storage 111, an OS 113, and a data reading unit 152 is provided. This embodiment is different from the first embodiment, and other than that is the same as the first embodiment.

  The data storage medium 150 is a readable / writable and portable recording medium such as an MO. The data recording unit 151 performs processing for storing the data update operation information transmitted from the backup management unit 105 in the data storage medium 150. The data reading unit 152 of the backup machine 110 reads the update operation information of the data stored in the data storage unit 150, and performs the process of updating the data in the backup storage 111 through the OS 113 based on the read update operation information. .

  Next, operations performed when the application program 101 updates the data in the storage 103 will be described in detail with reference to the flowcharts of FIGS.

  First, when data update operation information for the storage 103 is issued from the application program 101 (S1001 in FIG. 10A), a data update interrupt occurs in the OS 102 or a data update library is called (S1002), and backup management is performed. Control is transferred to the unit 105. The backup management unit 105 analyzes the update operation information of the data issued by the application program 101 and collates with the backup condition information set in the backup condition table 104 (S1003). Then, if there is backup condition information that matches the update operation information of the data issued by the application program 101 (Yes in S1003), the backup management unit 105 notifies the data recording unit 151 of the update operation information (S1004). If there is no suitable backup condition information (No in S1003), the process of step S1004 is skipped, and the storage 103 is updated based on the update operation information through the OS 102 (S1005). Then, the data update result is notified to the application program 101 through the OS 102 (S1013, S1014). The application program 101 proceeds with processing according to the update result (S1015). On the other hand, the data recording unit 151 stores the data update operation information passed from the backup management unit 105 in the data storage medium 150 (S1500).

  In the process example of FIG. 10A, the storage update process in step S1005 is performed independently of the process in step S1500, and therefore the delay of the data update process in the computer 100 can be reduced. However, even if the data in the storage 103 is updated in the computer 100, a state in which no data is stored in the data storage medium 150 may occur. For this reason, in order to improve reliability, the storage update process in step S1005 may be performed after confirming the completion of the data storage process in step S1500. Alternatively, as shown in the flowchart of FIG. 10B, the result of the data storage process in step S1500 is notified from the data recording unit 151 to the backup management unit 105, and in the backup management unit 105, the storage update process and step in step S1005 are performed. It is determined whether or not both of the data storage processes in S1500 have succeeded (S1011). If only one of them has failed, the process S1012 for invalidating the update on the successful side is performed. May be. In order to implement the update invalidation processing, the data before the update of the storage 103 and the data before the update of the data storage medium 150 are stored in, for example, an internal storage managed by the OS 102, and the update is invalidated. In this case, it is possible to write back to the data before update stored in the internal memory.

  In the present embodiment, the backup storage 111 of the backup machine 110 is updated based on the update operation information of the data stored in the data storage medium 150. In this case, the data storage medium 150 is transported to the installation location of the backup machine 110 and connected to the backup machine 110. When the data storage medium 150 is connected, the data reading unit 152 sequentially reads data update operation information recorded on the data storage medium 150 and updates the data in the backup storage 111 via the OS 113. As a result, the data in the backup storage 111 is updated in the same manner as the storage 103, and the data is multiplexed. Note that the data storage medium 150 and the backup storage 111 may be the same, and in this case, data read processing is unnecessary.

  FIG. 11 shows an example of the backup condition table 104 used in the present embodiment. Unlike the backup condition tables C2000 to C2200 used in the first embodiment, the backup condition table C2300 in this example specifies localhost indicating the local computer 100 as the backup destination machine name, and the data storage medium 150 in the backup destination path. Specify the device name. The backup condition information R2301 specifies “/ home / usr / [am] *” as the backup source path name, the file is the directory / home / usr, and the first character of the file name is between a and m. Means that the file is saved to the device specified in / dev / od1. The backup condition information R2302 specifies “/ home / usr / [nz] *” as the backup source path name, the file is the directory / home / usr, and the first character of the file name is between n and z. If it is alphabetic, it means save to the device specified in / dev / od2.

  When the backup condition table C2300 as shown in FIG. 11 is used, the backup management unit 105 performs the update of the storage 103 to be updated by the application program 101 in the backup condition matching process S1003 of FIGS. 10-1 and 10-2. Backup condition information having a backup source path that matches the file path is searched from the backup condition table C2300. If there is backup condition information that matches the backup source path, the schedule column is referred to, and if it is set, backup condition information for which the schedule condition is true is extracted. Then, update operation information for updating is stored in the device specified as the backup destination path in the backup condition information in the same manner as the file in the storage 103.

  Instead of the backup condition table C2300 shown in FIG. 11, it is also possible to use a backup condition table having application program names and user names as backup conditions as shown in FIGS.

  Next, the effect of this embodiment will be described.

  According to the present embodiment, when the application program 101 updates the data in the storage 103, the update operation information for saving only the update data satisfying the preset backup condition in the backup storage 111 installed in the backup machine 110 is data. It can be stored in the storage medium 150. The reason is that the backup management table 105 is provided with a backup condition table 104 that holds at least one backup condition information. When the update operation information for the data in the storage 103 is issued from the application program 101 to the OS 102, the backup management unit 105 stores the update operation information. This is because the data update operation information of this time is recorded in the data storage medium 150 only when the backup condition information that is analyzed and matched is searched from the table 104 and the matching backup condition information is registered.

  Further, according to the present embodiment, since it is not necessary to connect the computer 100 and the backup machine 110 online via a network, a backup system can be constructed at a lower cost.

“Third embodiment”
Referring to FIG. 12, in the third embodiment of the present invention, one backup machine 140-1 is connected to the computer 100 through the network 201, and another backup machine 140-1 is connected to the backup machine 140-1 through the network 201. A plurality of backup machines are sequentially connected in a chain such that another backup machine (not shown) is connected to the backup machine 140-2 through the network 201, and In addition to the OS 113 and backup storage 111-1, 111-2 similar to the backup machine in FIG. 1, each backup machine 140-1, 140-2 includes a data transfer unit 120, backup condition tables 121-1, 121-2, and The first embodiment is different from the first embodiment in that the backup management unit 122 is provided. To differences. The backup machine at the end of the chain does not need a backup condition table and a backup management unit. The components of each backup machine generally have the following functions.

  The data transfer unit 120 communicates with the computer 100 or another backup machine via a network. When data update operation information is received from the connected computer 100 or another backup machine, the data transfer unit 120 issues the update operation information to the OS 113 of the own computer. Further, the data transfer unit 120 performs a process of transmitting the data update operation information notified from the backup management of the own computer to the next-stage backup machine.

  The backup condition tables 121-1 and 121-2 are tables that hold backup condition information of the backup storages 111-1 and 111-2, and describe at least one backup condition information. The format of the backup condition information is the same as the backup condition information in the backup condition table 104.

  The backup management unit 105 selects any of the data update operation information for the data in the backup storages 111-1 and 111-2 issued from the data transfer unit 120 to the OS 113 and held in the backup condition tables 121-1 and 121-2. The operation information matching the backup condition information is detected, and whether or not the data update operation information can be transferred to the backup machine back in the next stage is controlled according to the detection result.

  Next, of the operations of the present embodiment, the operation when the application program 101 updates the data in the storage 103 will be described. As a premise, regarding the data to be multiplexed, it is assumed that the data in the storage 103 and the data in the backup storages 111-1 and 111-2 are synchronized.

  In the computer 100, when a certain file update in the storage 103 by the application program 101 matches the backup condition information held in the backup condition table 104, the data transmission / reception unit 106 connects to the network as in the first embodiment. Data update operation information is transmitted to the backup machine 140-1 via 201.

  In the backup machine 140-1, the data transfer unit 120 issues the received data update operation information to the OS 113 to update the backup storage 111-1, and at that time, an interrupt of data update occurs in the OS 119 or data update is performed. The library is called and control is transferred to the backup management unit 122. The backup management unit 122 collates the backup condition information read from the backup condition table 121 with the data update operation information, and if the backup condition is met, notifies the data transfer unit 120 of the data update operation information, and the data transfer unit 120 The data update operation information is transmitted to another backup machine 140-2 through the network 202. Thereafter, the backup machine 140-2 performs the same operation as the backup machine 140-1, thereby performing backup in a chained manner. In this way, when performing chained processing, waiting for confirmation of backup storage update increases the processing delay of the backup source, and therefore, processing such as waiting for confirmation is omitted as in the processing example of FIG. 2-1. Is preferred.

  As described above, according to this embodiment, the same effects as those of the first embodiment can be obtained, and the computer 100 can perform a plurality of backups only by communication with one backup machine 110-1. It becomes possible, and the processing load of the computer 100 can be reduced.

  In this embodiment, each backup machine also includes a backup condition table and a backup management unit, so that a certain file in the storage 103 is stored in the backup machine 140-1 only, and another file is stored in the backup machine 140-1. Fine backup control such as storing backups in the two backup machines 140-1 and 140-2 becomes possible.

  In this embodiment, each backup machine is provided with a backup condition table and a backup management unit. However, the data update operation information received by the data transfer unit 120 is issued to the OS of the own computer, and at the same time, the next stage through the network. It can also be configured to transmit to a backup machine.

“Fourth embodiment”

  Referring to FIG. 13, in the fourth embodiment of the present invention, the computer 100 includes an encrypted data transmitting / receiving unit 130 and a backup condition table 104A instead of the data transmitting / receiving unit 106 and the backup condition table 104, and each backup device 110- 1 to 110-3 differ from the first embodiment in that an encryption data transmission / reception unit 131 is provided instead of the data transmission / reception unit 112, and the other points are the same as those in the first embodiment.

  The backup condition table 104A holds at least one backup condition information, and each backup condition information includes information on a file, an application program, or a user as a backup target condition. Although this point is the same as the backup condition table 104 in the first embodiment, the difference is that each backup condition information includes a description of processing contents to be applied to data to be sent to the backup machine. The processing contents include security indicating that encryption is performed, data compression, and other data conversion. This embodiment shows an example in which information about security is held.

  FIG. 14 shows an example of the contents of the backup condition table 104A. The backup condition table C2000 in this example holds two pieces of backup condition information R2401 and R2402. Each piece of backup condition information includes each item of security as an example of a backup source path, a backup destination machine name, a backup destination path, and processing contents. The former three items are the same as in the first embodiment. In the security, a secure communication method used for communication between the computer 100 and the backup machines 110-1 to 110-3 is specified.

  Specifically, the backup condition information R2401 indicates that when the application program 101 updates files under the directory / home / usr1 / secret /, the backup destination machine name busrv is backed up with the same file path. It shows that DES encryption communication is used as a secure communication method.

  In addition, when the application program 101 updates files under the / home / usr2 directory, the backup condition information R2402 is backed up to the backup destination machine name busrv with the same file path, and PKI Electronics is used as a secure communication method at that time. Indicates that signed communication is used.

  The operation of this embodiment is almost the same as that of the first embodiment. The difference is that the backup up management unit 105 acquires a secure communication method from the matched backup condition information and designates the secure communication method. And requesting the encrypted data transmitting / receiving unit 130 to transfer the data update operation information. In response to the transfer request from the backup management unit 105, the encrypted data transmission / reception unit 130 encrypts transmission data using a designated secure communication method and transmits the encrypted transmission data to the backup servers 110-1 to 110-3. The encrypted data transmission / reception unit 130 performs decryption when receiving the encryption from the backup machines 110-1 to 110-3. On the other hand, the encrypted data transmission / reception unit 131 in each of the backup machines 110-1 to 110-3 decrypts the encrypted data at the time of reception, and encrypts and transmits the encrypted data using the same secure communication method at the time of transmission.

  As described above, according to the present embodiment, the same effect as that of the first embodiment can be obtained, and the communication between the computer 100 and the backup machines 110-1 to 110-3 is encrypted. Even when an open network such as the Internet is used, backup data can be safely stored in a remote backup storage. In addition, encryption is not performed uniformly, but there is an effect that it is possible to select whether or not encryption can be performed and a secure communication method can be selected for each file, application, or user specified by the backup condition information.

"Fifth embodiment"
Referring to FIG. 15, in the fifth embodiment of the present invention, a plurality of backup source computers 100-1 and 100-2 are connected through a network 201 to a plurality of backup devices 110-1 to 110-3. And the point that each backup machine 110-1 to 110-3 includes machine management tables 115-1 to 115-3 and a machine management unit 114, is different from the first embodiment.

  The machine management tables 115-1 to 115-3 hold at least one machine management information. Each machine management information includes a backup source machine name, a backup source path, a backup destination path, and filter information. The backup destination path is required when changing the path information on the backup machine side. The filter information is information that specifies a filter program that performs conversion of Japanese codes, and is necessary when backing up information that has passed through some filter.

  FIG. 16 shows an example of the machine management table 115-1. The machine management information R2501 should be converted to the path "/ server1 / usr1" and saved when data backed up with the path "/ home / usr1" is received from the backup source computer with the machine name server1. Indicates. That is, /home/usr1/file.txt of server1 is backed up as /server1/usr1/file.txt in the backup machine 110-1.

  On the other hand, the machine management information R2502 is converted to “/ server2 / usr1” when receiving data to be backed up by the path “/ home / usr1” from the backup source computer with the IP address 192.168.0.200, and further “nkf” -w ”means saving through the Japanese code conversion filter program.

  The same contents as the machine management table 115-1 of the backup machine 110-1 are registered in the machine management tables 115-2 and 115-3 of the other backup machines 110-2 and 110-3.

  The machine management unit 114 of each of the backup machines 110-1 to 110-3 uses the data update operation information path received from the data transmission / reception unit 112 from the backup source computers 100-1 and 100-2, and the machine conforming thereto. Conversion is performed based on the backup destination path in the machine management information held in the management tables 115-1 to 115-3, and when the filter information is designated, the designated filter program is executed.

  In the operation of this embodiment, the operations of the backup source computers 100-1 and 100-2 are the same as those of the first embodiment, and the difference is in the plurality of backup servers 110-1 to 110-3. Is the action. The operation will be described below taking the backup machine 110-1 as an example. The other backup machines 110-2 and 110-3 perform the same operation.

  When the data transmission / reception unit 112 receives the data update operation information from the backup source computers 100-1 and 100-2, it transmits it to the machine management unit 114. The machine management unit 114 refers to the machine management table 115-1 and collates with the data update operation information to determine a data storage destination path and a necessary filter program. For example, if the IP address of the computer 100-1 that is the transmission source of the data update operation information received this time is 192.168.0.200 and the backup source path specified by the received information is / home / usr1, the machine management information R2502 in FIG. Therefore, the machine management unit 114 converts the path name to be saved based on the path name / server2 / usr1 specified as the backup destination path, and if a filter program such as Japanese code conversion is specified. Save through that filter.

  As described above, according to the present embodiment, the same effects as those of the first embodiment can be obtained, and the system construction cost for collecting and storing data of a plurality of backup source computers in a plurality of backup machines can be obtained. Can be reduced. In addition, the storage path can be freely changed on each backup machine side, and processing by an arbitrary filter program such as conversion of Japanese codes can be performed on the backup data.

“Sixth embodiment”
Referring to FIG. 17, in the sixth embodiment of the present invention, the data transmission / reception unit 106 of the computer 100 is omitted, and instead of the backup machines 110-1 to 110-3 being connected via the network 201, a backup storage is provided. 111 is connected, and the backup management unit 105 is different from the first embodiment in that the backup data of the storage 103 is stored in the local backup storage 111.

  The backup storage 111 is an external storage device configured by a hard disk device or the like, and stores a copy of data stored in the storage 103 in file units. The backup storage 111 does not have to have the same capacity as the storage 103, and specifications (specifications such as hard disk cylinders, tracks, sectors, and headers) may be different.

  Next, operations performed when the application program 101 updates data in the storage 103 among the operations of the present embodiment will be described in detail with reference to the flowcharts of FIGS.

  When data update operation information for the storage 103 is issued from the application program 101 (S1001 in FIG. 18A), a data update interrupt occurs in the OS 102 or a data update library is called (S1002), and the backup management unit 105 Control is transferred to. The backup management unit 105 analyzes the update operation information of the data issued by the application program 101 and collates with the backup condition information set in the backup condition table 104 (S1003). If the backup condition information that matches the update operation information of the data issued by the application program 101 exists (Yes in S1003), the backup management unit 105 uses the same storage 103 and backup storage 111 based on the update operation information. If there is no suitable backup condition information (No in S1003), only the storage 103 is updated based on the update operation information (S1005). Then, the data update result is notified to the application program 101 through the OS 102 (S1013, S1014). The application program 101 proceeds with processing according to the update result (S1015).

  In the processing example of FIG. 18A, the storage update process in step S1005 is performed independently of the backup storage update process in step S104A. Therefore, even if the data in the storage 103 is updated, the backup storage 111 is not updated. A condition can occur. For this reason, in order to improve reliability, the storage update process in step S1005 may be performed after confirming the completion of the backup storage update process in step S1004A. Alternatively, as shown in the flowchart of FIG. 18-2, the backup management unit 105 determines whether both the storage update process in step S1005 and the backup storage update process in step S1004A have succeeded (S1011). If only one of them fails, the process S1012 for invalidating the update on the successful side may be performed. In order to realize the update invalidation processing, the data before the update of the storage 103 or the data before the update of the backup storage 111 is stored in, for example, an internal storage managed by the OS 102, and the update is invalidated. Is possible by writing back to the pre-update data stored in the internal memory.

  FIG. 19 shows an example of the backup condition table 104 used in the present embodiment. Unlike the backup condition tables C2000 to C2200 used in the first embodiment, the backup condition table C2600 in this example specifies localhost indicating the own computer 100 as the backup destination machine name. The backup condition information R2601 specifies "/ home / usr / [am] *" as the backup source path name, the file is the directory / home / usr, and the first character of the file name is between a and m Means that the file is stored in a file under / bp1 / usr in the backup storage 111. The backup condition information R2302 specifies “/ home / usr / [nz] *” as the backup source path name, the file is the directory / home / usr, and the first character of the file name is between n and z. In the case of the alphabet, it means saving in a file under / bp1 / use of the backup storage 111.

  When the backup condition table C2600 as shown in FIG. 19 is used, the backup management unit 105 performs the update of the storage 103 to be updated by the application program 101 in the backup condition matching process S1003 of FIGS. 18-1 and 18-2. The backup condition information having a backup source path that matches the file path is searched from the backup condition table C2600. If there is backup condition information that matches the backup source path, the schedule column is referred to, and if it is set, backup condition information for which the schedule condition is true is extracted. Then, the same update as the file in the storage 103 is performed on the file in the path of the backup storage 111 specified as the backup destination path in the backup condition information.

  Instead of the backup condition table C2300 shown in FIG. 19, it is also possible to use a backup condition table having application program names and user names as backup conditions as shown in FIGS.

  According to the present embodiment, when the application program 101 updates the data in the storage 103, only update data that satisfies a preset backup condition can be stored in the local backup storage 111.

  Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various other additions and modifications can be made. Also, each component of the backup management unit, data transmission / reception unit, data recording unit, encrypted data transmission / reception unit in the backup source computer of the present invention, data transmission / reception unit, data reading unit, data transfer unit, backup management unit in the backup destination computer Each component of the encrypted data transmission / reception unit and the machine management unit can be realized by a computer and a program. The program is provided by being recorded on a computer-readable recording medium such as a magnetic disk or a semiconductor memory, and is read by the computer at the time of starting up the computer and the operation of the computer is controlled to thereby control the computer. Act as an element.

It is a block diagram of a 1st embodiment of the present invention. It is a flowchart which shows the process example at the time of the data update in the 1st Embodiment of this invention. It is a flowchart which shows another example of a process at the time of the data update in the 1st Embodiment of this invention. It is a flowchart of the process which takes the initial stage operation | movement in the 1st Embodiment of this invention. It is a block diagram explaining an example of the recovery operation in the 1st Embodiment of this invention. It is a flowchart which shows the other example of the recovery operation in the 1st Embodiment of this invention. It is a figure which shows an example of the backup condition table used in the 1st Embodiment of this invention. It is a figure which shows another example of the backup condition table used in the 1st Embodiment of this invention. It is a figure which shows another example of the backup condition table used in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the 2nd Embodiment of this invention. It is a flowchart which shows the example of a process at the time of the data update in the 2nd Embodiment of this invention. It is a flowchart which shows another example of a process at the time of the data update in the 2nd Embodiment of this invention. It is a figure which shows an example of the backup condition table used in the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the 3rd Embodiment of this invention. It is a block diagram which shows the structure of the 4th Embodiment of this invention. It is a figure which shows an example of the backup condition table used in the 4th Embodiment of this invention. It is a block diagram which shows the structure of the 5th Embodiment of this invention. It is a figure which shows an example of the machine management table used in the 5th Embodiment of this invention. It is a block diagram which shows the structure of the 6th Embodiment of this invention. It is a flowchart which shows the example of a process at the time of the data update in the 6th Embodiment of this invention. It is a flowchart which shows another example of a process at the time of the data update in the 6th Embodiment of this invention. It is a figure which shows an example of the backup condition table used in the 6th Embodiment of this invention.

Explanation of symbols

100, 100-1, 100-2 ... computer 101 ... application program 102, 113, 119 ... OS
103 ... Storage 104, 121-1, 121-2 ... Backup condition table 105, 122 ... Backup management unit 106, 112 ... Data transmission / reception unit 110, 110-1, 110-2, 110-3 ... Backup machine 111, 111- 1, 111-2, 111-3, 123-1, 123-2 ... Backup storage 114 ... Machine management unit 115-1, 115-2, 115-3 ... Machine management table 120 ... Data transfer unit 130, 131 ... Encryption Data transmission / reception unit 150... Data storage medium 151... Data recording unit 152.

Claims (18)

A computer system in which a master computer and one or more slave computers are connected via a network,
The slave computer is a slave operating system, a backup storage, and a slave that updates the backup storage via the slave operating system based on update operation information for the data of the backup storage transmitted from the master computer. Side data transmitting and receiving unit,
The master computer includes a master operating system, a storage for storing data, an application program for updating data in the storage through the master operating system, a backup condition table for storing backup condition information including file information, The master side data transmitting / receiving unit for communicating with the slave side data transmitting / receiving unit, the update operation information for the storage data issued from the application program, the update operation information including file information for specifying the update file Obtaining via the master side operating system, updating the storage through the master side operating system based on the obtained update operation information, and the obtained update operation It is determined whether backup condition information including file information matching the file information included in the information is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that the backup is to be performed. The update operation information for performing the same update as the storage update based on the acquired update operation information to the backup storage is transmitted to the one or more slave computers via the master side data transmitting / receiving unit. A computer system comprising a backup management unit. A computer system in which a master computer and one or more slave computers are connected via a network,
The slave computer is a slave that updates the backup storage via the slave side operating system based on update operation information for the backup storage data transmitted from the master computer, the slave side operating system, the backup storage, and Side data transmitting and receiving unit,
The master computer includes a master side operating system, a storage for storing data, an application program for updating data in the storage through the master side operating system, and a backup condition table for storing backup condition information including application program information; A master-side data transmission / reception unit for communicating with the slave-side data transmission / reception unit, and update operation information for the storage data issued from the application program and including file information for specifying an update file. The storage is acquired through the master side operating system, and the storage is updated through the master side operating system based on the acquired update operation information. It is determined whether backup condition information including application program information that matches the application program information of the application program that issued the acquired update operation information is stored in the backup condition table, and the corresponding backup condition information is stored. If it is determined that backup is to be performed, update operation information for performing the same update as the storage update based on the acquired update operation information to the backup storage is transmitted via the master side data transmitting / receiving unit. And a backup management unit for transmitting to the one or more slave computers. A computer system in which a master computer and one or more slave computers are connected via a network,
The slave computer is a slave operating system, a backup storage, and a slave that updates the backup storage via the slave operating system based on update operation information for the data of the backup storage transmitted from the master computer. Side data transmitting and receiving unit,
The master computer includes a master operating system, a storage for storing data, an application program for updating data in the storage through the master operating system, a backup condition table for storing backup condition information including user information, The master side data transmitting / receiving unit for communicating with the slave side data transmitting / receiving unit, the update operation information for the storage data issued from the application program, the update operation information including file information for specifying the update file Obtaining via the master side operating system, updating the storage through the master side operating system based on the obtained update operation information, and obtaining the obtained update operation information It is determined whether backup condition information including user information that matches the user information of the user who started the application program that has issued is stored in the backup condition table, and if the corresponding backup condition information is stored It is determined that backup is to be performed, and update operation information for performing an update similar to the update of the storage based on the acquired update operation information to the backup storage is performed via the master side data transmitting / receiving unit. A computer system comprising a backup management unit for transmitting to the slave computer.   The backup condition information further includes information related to a slave computer serving as a backup destination, and the backup management unit applies to the slave computer specified by the information related to the slave computer serving as a backup destination included in the backup condition information. 4. The computer system according to claim 1, wherein the update operation information is transmitted.   The backup condition information further includes information related to a backup schedule, and the backup management unit checks whether the schedule included in the corresponding backup condition information is established, and if the schedule is established, the backup is performed. 5. The computer system according to claim 4, wherein the computer system is determined to be performed, and if the schedule is not established, it is determined that backup is not performed.   6. The computer system according to claim 1, wherein the computer system is configured to synchronize the update of the storage by the master computer and the update of the backup storage by the slave computer corresponding to the update. . A master computer and one or more slave computers are connected via a network, and the slave computer includes update operation information for the data in the backup storage transmitted from the slave operating system, the backup storage, and the master computer. Based on the slave-side operating system, the master-side data transmission / reception unit updates the backup storage, and the master computer stores the master-side operating system, storage for storing data, and data in the storage as the master Application program to be updated through the operating system, backup condition table for storing backup condition information including file information, and the slave data transmission / reception unit A backup method in a computer system comprising a master-side data transceiver for communicating,
The backup management unit of the master computer obtains, via the master side operating system, update operation information including file information for specifying the update file, which is update operation information for the storage data issued from the application program. The storage condition is updated through the master side operating system based on the acquired update operation information, and backup condition information including file information matching the file information included in the acquired update operation information is stored in the backup condition table. It is determined whether the backup condition information is stored, and if the corresponding backup condition information is stored, it is determined that the backup is performed, and the same update as the storage update based on the acquired update operation information is performed. Backup method characterized by the updating operation information through the master-side data transceiver to transmit to the one or more slave computers for performing relative storage. A master computer and one or more slave computers are connected via a network, and the slave computer includes update operation information for the data in the backup storage transmitted from the slave operating system, the backup storage, and the master computer. Based on the slave-side operating system, the master-side data transmission / reception unit updates the backup storage, and the master computer stores the master-side operating system, storage for storing data, and data in the storage as the master An application program to be updated through the operating system, a backup condition table for storing backup condition information including application program information, and the slave A backup method in a computer system comprising a master-side data transceiver for communicating with the side data transceiver,
The backup management unit of the master computer obtains, via the master side operating system, update operation information including file information for specifying the update file, which is update operation information for the storage data issued from the application program. Backup condition information including application program information that matches the application program information of the application program that issued the acquired update operation information, while updating the storage through the master side operating system based on the acquired update operation information Is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that the backup is performed, and the acquired update is performed. Backup the same updates and update of the storage system based on the operation information and transmits the updated operation information for performing with respect to the backup storage to said one or more slave computer via the master-side data transceiver Method. A master computer and one or more slave computers are connected via a network, and the slave computer includes update operation information for the data in the backup storage transmitted from the slave operating system, the backup storage, and the master computer. Based on the slave-side operating system, the master-side data transmission / reception unit updates the backup storage, and the master computer stores the master-side operating system, storage for storing data, and data in the storage as the master An application program to be updated through the operating system, a backup condition table for storing backup condition information including user information, and the slave data transmitter / receiver A backup method in a computer system comprising a master-side data transmitting and receiving unit for signal,
The backup management unit of the master computer obtains, via the master side operating system, update operation information including file information for specifying the update file, which is update operation information for the storage data issued from the application program. A backup that includes user information that matches the user information of the user who started the application program that issued the acquired update operation information, while updating the storage through the master side operating system based on the acquired update operation information It is determined whether or not condition information is stored in the backup condition table. If the corresponding backup condition information is stored, it is determined that backup is to be performed, and the scan based on the acquired update operation information is determined. Backup method characterized by transmitting an update operation information for performing the same updates and update of storage to the backup storage to said one or more slave computer via the master-side data transceiver.   The backup condition information further includes information related to a slave computer serving as a backup destination, and the backup management unit is configured for a slave computer specified by information related to a slave computer serving as a backup destination included in the backup condition information. The backup method according to claim 7, wherein the update operation information is transmitted.   The backup condition information further includes information regarding a backup schedule, and the backup management unit checks whether the schedule included in the corresponding backup condition information is established, and if the schedule is established, the backup is performed. The backup method according to claim 10, wherein the backup method is determined to be performed, and if the schedule is not established, it is determined that the backup is not performed.   The backup method according to any one of claims 7 to 11, wherein the update of the storage by the master computer is synchronized with the update of the backup storage by the slave computer corresponding to the update. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; A computer connected via a network to one or more slave computers equipped with
Master side operating system, storage for storing data, application program for updating data in the storage through the master side operating system, backup condition table for storing backup condition information including file information, and slave side data transmission / reception A master-side data transmission / reception unit for communicating with the storage unit, and update operation information for the storage data issued from the application program, the update operation information including file information for specifying the update file being transmitted to the master-side operating system. And updating the storage through the master side operating system based on the acquired update operation information and the file included in the acquired update operation information. It is determined whether backup condition information including file information that matches the information is stored in the backup condition table, and if the corresponding backup condition information is stored, it is determined that a backup is performed, and the acquired A backup management unit for transmitting update operation information for performing an update similar to the update of the storage based on update operation information to the backup storage to the one or more slave computers via the master-side data transmission / reception unit; A computer characterized by comprising. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; A computer connected via a network to one or more slave computers equipped with
Master side operating system, storage for storing data, application program for updating data in the storage through the master side operating system, backup condition table for storing backup condition information including application program information, and slave side data A master-side data transmission / reception unit for communicating with the transmission / reception unit; and update operation information for the storage data issued from the application program, the update-operation information including file information for specifying an update file, the master-side operating system And updating the storage through the master side operating system based on the acquired update operation information and the acquired update operation information. It is determined whether backup condition information including application program information that matches the application program information of the application program that has issued is stored in the backup condition table. If the corresponding backup condition information is stored, backup is performed. The update operation information for performing update similar to the update of the storage based on the acquired update operation information to the backup storage is determined via the master-side data transmission / reception unit. A computer comprising: a backup management unit that transmits to a slave computer. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; A computer connected via a network to one or more slave computers equipped with
Master side operating system, storage for storing data, application program for updating data in the storage through the master side operating system, backup condition table for storing backup condition information including application program information, and slave side data A master-side data transmission / reception unit for communicating with the transmission / reception unit; and update operation information for the storage data issued from the application program, the update-operation information including file information for specifying an update file, the master-side operating system And updating the storage through the master side operating system based on the acquired update operation information and the acquired update operation information. It is determined whether backup condition information including user information that matches the user information of the user who started the application program that has issued is stored in the backup condition table, and if the corresponding backup condition information is stored It is determined that backup is to be performed, and update operation information for performing an update similar to the update of the storage based on the acquired update operation information to the backup storage is performed via the master side data transmitting / receiving unit. A computer comprising: a backup management unit that transmits data to the slave computer. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; Including a master side operating system, a storage for storing data, an application program for updating data in the storage through the master side operating system, and file information. A computer equipped with a backup condition table for storing backup condition information
A master side data transmitting and receiving unit for communicating with the slave side data transmitting and receiving unit;
Update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, is acquired via the master side operating system, and based on the acquired update operation information Updating the storage through the master side operating system and determining whether backup condition information including file information matching the file information included in the acquired update operation information is stored in the backup condition table. If the relevant backup condition information is stored, it is determined that the backup is performed, and update operation information for performing the same update as the storage update based on the acquired update operation information on the backup storage A backup management unit to be transmitted through the master-side data transceiver to the one or more slave computer,
Program to make it function. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; A master side operating system, a storage for storing data, an application program for updating data in the storage through the master side operating system, and application program information. A computer having a backup condition table for storing backup condition information including
A master side data transmitting and receiving unit for communicating with the slave side data transmitting and receiving unit;
Update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, is acquired via the master side operating system, and based on the acquired update operation information Backup condition information including application program information that matches the application program information of the application program that issued the acquired update operation information is stored in the backup condition table. And if the relevant backup condition information is stored, it is determined that the backup is performed, and the storage update based on the obtained update operation information is determined. A backup management unit that transmits an update operation information for performing like updates to the backup storage to said one or more slave computer via the master-side data transceiver,
Program to make it function. A slave-side operating system, a backup storage, and a slave-side data transmission / reception unit that updates the backup storage via the slave-side operating system based on update operation information for the data in the backup storage transmitted from another computer; A master side operating system, a storage for storing data, an application program for updating data in the storage through the master side operating system, and application program information. A computer having a backup condition table for storing backup condition information including
A master side data transmitting and receiving unit for communicating with the slave side data transmitting and receiving unit;
Update operation information for the storage data issued from the application program, including update information including file information for specifying an update file, is acquired via the master side operating system, and based on the acquired update operation information The storage condition is updated through the master side operating system, and backup condition information including user information that matches the user information of the user who started the application program that issued the acquired update operation information is stored in the backup condition table. If the relevant backup condition information is stored, it is determined that backup is to be performed, and an update similar to the storage update based on the obtained update operation information is performed. A backup management unit that transmits an update operation information for performing relative flop storage to said one or more slave computer via the master-side data transceiver,
Program to make it function.

Images