JP2001125815A - Back-up data management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system for realizing the unitary management of back-up data without depending on the format of data to be backed-up in a system environment in which plural OSs in different types coexist. SOLUTION: A log file 41 stores log records having a data writing time, a data storing place, and data contents to be recorded corresponding to a write command issued by hosts 4 and 5. A log capturing part 20 is started by a log capture controlling part 7 so that the recording of the log file 41 can be started for a designated device at the designated time. A data resorting part 21 is started by a data restoration controlling part 8 so that the data of the designated device can be updated and restored based on the log records after the designated time. The data transfer program of a data transferring part 9 is executed by a WS6, and the movement of data between a magnetic disk sub-system 2 and a storage sub-system 3 is repeated at the designated time.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for managing backup data of a storage device using a rewritable storage medium, and more particularly to a mechanism for collecting a log of write data in a storage subsystem and an old data and log. The present invention relates to a backup data management system for providing a mechanism for restoring new data from a backup data and controlling these mechanisms to control backup data storage and data restoration.

[0002]

2. Description of the Related Art In a storage device using a rewritable storage medium such as a magnetic disk device, the data access speed tends to increase, and the storage capacity tends to increase year by year.
Furthermore, in recent years, a highly reliable storage device employing the disk array technology has been realized, and a system in which an external storage device of a computer is positioned at the center of the system and stored information is shared by many computers is becoming established. .

When an external storage device is installed at the center of the system, data backup for data security becomes an increasingly important technology. The disk array technology and the mirroring technology increase the reliability of the device itself, but do not completely realize data security. Therefore, it is still necessary to back up data in a secondary storage device such as an optical disk device having a highly reliable storage medium even in a magnetic disk device employing such a technique.

In a mainframe computer, software that operates on the operating system (OS) and backs up data of the entire system is available. The database management system has a function of backing up a database to be managed. On the other hand, in general-purpose computers other than mainframes, the format of backup data is the OS, database management system,
It depends on the application software,
It is difficult to output backup data having a common format in a system in which a plurality of OSs, database software, and applications coexist.

FIG. 5 is a diagram for explaining a conventional backup data creation situation. Two host computers 4 and 5 having different OSs and file systems 91 and 92 respectively connect devices 11 and 14 in the magnetic disk subsystem 90.
Are acquired as the backup files 93 and 94, there are two types of files having different data formats of the backup files 93 and 94 and having no compatibility. It is difficult to construct a backup data management system that manages the backup file 93 and the backup file 94 and can reproduce past data.

[0006]

As described above, according to the prior art, the backup data of a widely used general-purpose computer has a data format dependent on the OS and the like, and the backup data having a common format is output. However, it is difficult to provide software for centrally managing the output backup data. Further, when data in a file is changed or deleted by mistake due to a human operation error or the like, it cannot be recovered without backup data of the file. Furthermore, the proof of rewritable storage media such as magnetic disks has not been confirmed,
If data at a past point in time can be reproduced, there is a possibility that such a storage medium will be recognized as a storage device having evidential properties.

An object of the present invention is to provide a system for centrally managing backup data without depending on the format of the data to be backed up. Another object of the present invention is to provide a backup data management system capable of restoring data lost due to an operation error or the like. Another object of the present invention is to provide a backup data management system capable of reproducing data at a certain point in the past.

[0008]

According to the present invention, there is provided a first storage subsystem for holding data accessible from a host device on a rewritable storage medium, and a backup data of the data of the first storage subsystem. A second storage subsystem for storing the data, wherein the system obtains a log of a data writing time, a data storage location, and a data content requested by a higher-level device; A first storage subsystem having means for updating and restoring data with logs collected after a designated time, and means for activating the means for collecting the logs and the means for restoring the data according to time designation; Means for relaying the movement of data between the first storage subsystem and the second storage subsystem when a specified time is reached; Wherein the backup data management system and a second storage subsystem to perform the data write and data read as one destination and migration source data of the storage subsystem.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a configuration diagram of a backup data management system according to the present embodiment. The system is host 4,
Host 5, at least one workstation (W
S) 6, a magnetic disk subsystem 2, a storage subsystem 3, a monitoring device 1, and a transmission line 32 connecting the devices;
And a LAN (Local Area Network) 33.
The host 4 and the host 5 are computers equipped with different types of OSs, respectively, and are higher-level devices of the magnetic disk subsystem 2 and the storage subsystem 3. The magnetic disk subsystem 2 is a subsystem of a magnetic disk having a disk array device. Storage subsystem 3
Is a subsystem of a storage device using a magnetic tape, a magnetic disk, an optical disk, a magneto-optical disk or the like as a medium.
The monitoring device 1 is a computer that instructs the collection of a write log of the magnetic disk device of the magnetic disk subsystem 2 and monitors the progress of data migration between the magnetic disk subsystem 2 and the storage subsystem 3. Functions as a server computer. The WS 6 is a computer that is a higher-level device of the magnetic disk subsystem 2 and the storage subsystem 3.

The host 4, the host 5, the WS 6, the magnetic disk subsystem 2, and the storage subsystem 3 are connected by a high-speed data transfer network including a transmission line 32 and a hub 31. Such a system in which the storage subsystems are connected to each other by a high-speed dedicated data transfer network is called a SAN (Storage Area Network). A slower LA between the host 4, the host 5, the WS 6, the magnetic disk subsystem 2 and the monitoring device 1.
Connected by N33. The storage subsystem 3 may be connected to the LAN 33. The network centered on the LAN 33 is a control network, and is a network for communication when the monitoring device 1 issues a command to the magnetic disk subsystem 2 and the WS 6 and monitors the progress. Although not related to the present invention, in the configuration of the present embodiment, the hosts 4 and 5 and the WS 6 are connected to the LAN 33.
And the hosts 4 and 5 can also be used as a server-client system functioning as a WS6 server. If the processing performance is satisfied, a personal computer may be used instead of the WS.

The magnetic disk subsystem 2 includes a transmission path 3
, A plurality of devices 11 to 19, a control unit 22, a log collecting unit 20, and a data restoring unit 2.
1 has a module for performing each control and processing. The device 19 stores a log file 41 for recording a history of writing data to any device. The control unit 22 has an interface for communicating with the hosts 4, 5, and WS6 via the transmission line 32, interprets commands sent from the hosts 4, 5, and WS6, and performs communication between these devices and the devices 11 to 19. Control data transfer. The control unit 22 has an interface that communicates with the monitoring device 1 and the WS 6 via the LAN 33, and stores commands and control information via the LAN 33 in the log collection unit 20 or the data restoration unit 2
1 and outputs the information output by the log collection unit 20 or the data restoration unit 21 via the LAN 33 to the monitoring device 1 or WS6.
To send to. The log collection unit 20 is provided for the monitoring device 1 or the WS6
And records the history of write commands sent from the hosts 4 and 5 to the devices 11 to 19 in the log file 41. The data restoration unit 21 is activated by a command from the monitoring device 1 or the WS 6, refers to the log file 41 corresponding to any of the devices 11 to 19, and converts the data of the designated device into the log data of the corresponding log file 41. Update by. The LBA table 25 is a table accessed by the data restoration unit 21 in the second embodiment, and stores whether or not each block on the device has been updated. It is possible to provide a plurality of magnetic disk subsystems 2 in the system.

The storage subsystem 3 has ports connected to the host 4, the host 5, and the WS 6 via the transmission path 32, respectively, and has a plurality of devices 26 to 30. The control unit of the storage subsystem 3 has an interface for communicating with the hosts 4, 5, and WS6 via the transmission path 32, interprets commands sent from the hosts 4, 5, and WS6, and interprets these devices and devices 26 to 30 to control data transfer. The device 30 stores a log file 6 which is a backup copy of the log file 41 for a certain time period.
1 is stored. Devices 26 to 30 are devices 11 to
Each of them is also called an archive device because it is used to store, ie, archive, 19 backup data. The log file 61 is also called an archive log file. It is possible to provide a plurality of storage subsystems 3 in the system. The storage subsystem 3 may include the data restoration unit 21. When the storage subsystem 3 has the data restoration unit 21, the storage subsystem 3 is also connected to the LAN 33, and the control unit of the storage subsystem 3 operates in the same manner as the control unit 22 of the magnetic disk subsystem 2. 21 has an interface for communicating with the monitoring device 1 and the WS 6.

The processing unit of the monitoring device 1 stores programs of a log collection control unit 7, a data restoration control unit 8, a data migration unit 9, and a status display unit 10 in addition to the OS, and executes these programs. The monitoring information table 24 is a table for storing information indicating the progress of various processes. The processing device of the monitoring device 1 connects the display device 34 and the input device 35. The log collection control unit 7 specifies a device, a log file, and the like to be subjected to log collection of the magnetic disk subsystem 2, and specifies the log collection unit 20 via the LAN 33.
Start The data restoration control unit 8 designates a device to be subjected to data restoration, a log file to be referred to for data restoration, and the like.
Start The data migration unit 9 is a program that relays the movement of data or log files between the magnetic disk subsystem 2 and the storage subsystem 3.
When the WS 6 that executes data migration via the LAN 5, the source subsystem, the source device, the destination subsystem, the destination device, and the like are specified, the control information set to the specified WS 6 via the LAN 33 is transmitted. Send and instruct execution of data migration. Upon receiving this command, the WS 6 moves the designated source data or log file to the destination. The progress status of the log collection control unit 7 and the data restoration control unit 8 and the progress status of the data migration unit 9 based on the report from the WS 6 are recorded in the monitoring information table 24. The status display unit 10 edits information in the monitoring information table 24 according to an instruction from the input device 35 and displays the information on the display device 34.

The hosts 4, 5 and WS 6 access the devices 11 to 19 and the devices 26 to 30 by designating a common target ID and LUN (Logical Unit Number). The device specified by the target ID and the LUN is a logical device that can be seen from the hosts 4, 5 and the WS 6, and has a common storage capacity. The magnetic disk subsystem 2 and the storage subsystem 3 respectively convert the received target ID and LUN of the logical device into physical devices in the subsystem and addresses of storage areas in the physical device. In this specification, a device means a logical device unless otherwise specified.

Devices 11 to 19 and devices 26 to 3
Each data area in 0 is a fixed-length block (for example, 51
(2 byte area), and each block has an LBA (Logical Block Address).
The address called is given. LBA is a block number starting from 0 in each device. The hosts 4 and 5 and the WS 6 issue read / write commands specifying LBA and the number of blocks for devices in the magnetic disk subsystem 2 and the storage subsystem 3, respectively.
Access data in each device in block units.

FIG. 2 is a diagram showing the data format of the log file 41. The log file 41 is created for each device. Each log record of the log file 41 records the history of data added / updated by each write command, and has a time stamp, a start LBA, the number of blocks, and contents. “Timestamp” indicates the time of data writing by the write command, “Start LBA” indicates the data storage location, the address of the first block added / updated, “Number of blocks” indicates the number of blocks to be added / updated, “Content” indicates the content of the added / updated data for the number of target blocks. "Time stamp"
The data write time is the time at which data is written to a cache memory (not shown) or the time at which the data is written,
The time may be any time synchronized with the data writing, such as the time at which data is written to the physical device or the time at which the data is written, or the time at which a data write end signal is transmitted to the hosts 4 and 5. Each record of the log file 41 and the host 4,
5 corresponds to the write command issued. 51, 52
And 53 indicate a specific time as a time point. For example, time point 5
1 indicates a time between the time indicated by the time stamp of the log record 42 and the time indicated by the time stamp of the log record 43. The log file 61 has the same data format.

The log collection unit 20 can execute a plurality of the same log collection programs in the form of multiple programming. Each log collection program is responsible for collecting a log for one device specified by the log collection control unit 7, and the log device (for example, The log record is recorded in the log file 41 corresponding to the device in the device 19). The log collection unit 20 reports the progress of log collection to the log collection control unit 7.
Notify. The log collection control unit 7 records the progress in the monitoring information table 24.

The data restoration section 21 is provided with a data restoration control section 8.
File for storing a device to be restored by the data and a log record for the device
1 and when the restoration start time is designated and started, the "time stamp" of the designated log file 41 is searched, and the log records having the time stamps after the designated restoration start time are sequentially referred to and designated. Add / update data for the restored device and restore data up to the restoration end time. The data restoration unit 21 notifies the data restoration control unit 8 of the progress of data restoration. The data restoration control unit 8 records the progress in the monitoring information table 24.

The log collection control unit 7 is connected to the log collection unit 2 of the target magnetic disk subsystem 2 via the LAN 33.
Send a command to 0 to start log collection. This command specifies the number of the device to be collected, the number of the device that stores the log file, the name of the log file, the start and end times of log collection, and the size (storage capacity) of the log file. When the specified collection start time comes, the log collection unit 20 receives a write command to the log collection target device thereafter, and records a log record in the log file 41 of the specified log device at the time of data writing. When the file area of the log file is full of log records, wrap the log file at the beginning and continue recording log records until the specified end time is reached. The log collection control unit 7 can also stop the log collection by transmitting a log collection end command to the log collection unit 20.

The data restoration control unit 8 transmits a command to the data restoration unit 21 of the target magnetic disk subsystem 2 via the LAN 33 to start data restoration. This command specifies the number of the device to be restored, the number of the device that stores the log file, the name of the log file, the restoration start time and the end time. The data restoration unit 21 refers to the specified log file 41 of the specified log device, sequentially updates the data of the data restoration target device with the log record after the specified restoration start time, and terminates the specified restoration end. Continue the data restoration process until the time.

The WS 6 is connected to the monitoring device 1 via the LAN 33.
, The program of the log collection control unit 7 and / or the data restoration control unit 8 can be received, and the WS 6 can execute the processing of the log collection control unit 7 and / or the data restoration control unit 8. In this case, the WS 6 notifies the monitoring device 1 of the progress of the process received from the magnetic disk subsystem 2.

FIG. 3 is a flowchart showing the procedure of the data migration process. The data transfer unit 9 of the monitoring device 1 displays a guidance screen on the display device 34 and accepts input of transfer control information from the input device 35 (Step 71). The setting items are WS name, transfer source subsystem, transfer source device, transfer source log file name, transfer destination subsystem, transfer destination device, and transfer start time. The WS name is the name of the WS 6 that executes the migration process, the source subsystem is the identifier of the subsystem that stores the source data, the source device is the number of the device that stores the source data, and the destination subsystem is the data. The identifier of the subsystem to migrate to,
The migration destination device is the number of the device that receives the migrated data, and the migration start time is the time at which data migration starts. Here, the numbers of the transfer source and transfer destination devices are expressed by the target ID and LUN. The migration source log file name is the name of the log file to be the migration source, and is specified when the migration target is a log file. In the case of data migration from the magnetic disk subsystem 2 to the storage subsystem 3, the source subsystem is the magnetic disk subsystem 2, and the source devices are devices 11 to 19.
The destination subsystem is the storage subsystem 3, and the destination devices are the devices 26 to 30.
In the case of data migration from the storage subsystem 3 to the magnetic disk subsystem 2, the source and destination are reversed.
Next, the monitoring device 1 transmits the control information set via the LAN 33 to the WS 6 of the input WS name, that is, the WS 6 responsible for the data migration process (Step 72). A plurality of WSs 6 can be connected to the system, and the assigned WS 6 can be determined so as to distribute the load of the WS 6 processing.

The WS 6 receives the control information (Step 73) and starts a data migration program (Step 74). This data migration program is executed by the data migration unit 9
And stored by the monitoring device 1. If the WS 6 does not hold the copy, it can be obtained by downloading from the monitoring device 1 when the WS 6 logs in. WS
6 can execute a plurality of the same data migration programs in the form of multiple programming. Each data migration program executes an independent data migration process.

The activated data transfer program accesses the transfer source device of the specified transfer source subsystem via the transmission line 32 and reads the capacity of data occupying the device (step 75). If the source log file name is specified, the data size of the source log file is read. Next, the destination device of the destination subsystem is accessed, the free space of the device is read, and the two are compared to determine whether data migration is possible in terms of storage capacity (step 76). If data transfer is not possible, the monitoring device 1 is notified of the fact and the process is terminated. If data transfer is possible, wait until the specified transfer start time has come (step 77).

When the migration start time has been reached, the data migration program issues a READ command to the migration source device of the migration source subsystem via the transmission line 32 via the device driver held by the WS 6, and the data migration program issues a READ command. The data is read (step 78). Next, a WRITE command is issued to the destination device of the destination subsystem via the transmission path 32 via the device driver, and the read data of a predetermined number of blocks is written to the destination device (Step 79). Next, the progress status such as the number of blocks of data whose data has been migrated is transmitted to the monitoring device 1 (step 80). If the end of the transfer source device has been reached and the data transfer has not been completed (step 81 NO), the process returns to step 78 and repeats the above processing.

When the monitoring device 1 receives the data migration progress status (step 82), it stores the information in the monitoring information table 24 (step 83). When the data migration program of WS6 determines the end of data migration (YES in step 81), it notifies the monitoring device 1 of the end of data migration (step 84), and ends the processing. It should be noted that the processing of steps 71 and 72 can be executed by WS6. In this case, the WS 6 notifies the monitoring device 1 of the set control information.

In order for the data transfer unit 9 of the monitoring device 1 to back up the data of the devices of the magnetic disk subsystem 2, the transfer source subsystem controls the magnetic disk subsystem 2 and the transfer destination subsystem controls the storage subsystem 3. The information is transmitted to the WS 6, the data migration program of the WS 6 is activated, and the log acquisition control unit 7 activates the log acquisition unit 20 of the magnetic disk subsystem 2 to start the log acquisition of the data migration source device. . At this time, if the same time as the migration start time is designated as the start time, a log of the data written to the migration source device after the data migration starts can be collected. That is, the data update history for the data migration destination device (archive device) is recorded in the specified log file 41. In order to make a backup of the log file 41 of the magnetic disk subsystem 2, the data of the log file 41 can be transferred to the device 30 of the storage subsystem 3 and saved as an archive log file 61.

The data of the device of the magnetic disk subsystem 2 is migrated to the device of the storage subsystem 3, and the log of the source device of the magnetic disk subsystem 2 is collected in the log file 41 at the same time and the log file 41. When restoring the data of the migration destination device of the old storage subsystem 3 to the latest data after a lapse of time for storing the log records in the storage subsystem 3, the data of the old archive device of the storage subsystem 3 is first magnetically restored. Data is migrated to the device of the disk subsystem 2.
Next, when the data restoration control unit 8 of the monitoring apparatus 1 is started, and the same time when the log collection is started is set as the restoration start time, and the current time is set as the restoration end time, the data restoration unit 21 is started. The old archive data whose data has been migrated to 2 is updated to the latest data up to the specified current time. The data of the archive device of the storage subsystem 3 is transferred to the magnetic disk subsystem 2.
Of the storage subsystem 3 and transfer the archive log file 61 of the storage subsystem 3 to the magnetic disk subsystem 2
Then, the data restoration unit 21 is activated, and the archive data can be restored from the restoration start time to the restoration end time specified by the archive log file.

A specific example of log collection, data migration, and data restoration will be described below by exemplifying a specific device number and a specific point in time.

(1) Recovery of Data Due to Device Failure of Magnetic Disk Subsystem 2 For example, assume that data at device 17 is migrated to archive device 26 at time point 51 in FIG. 2 and backup data is stored. It is also assumed that the log record of the device 17 has been collected in the log file 41. Time point 53
Causes a device failure in the device 17,
It is assumed that the above data has become inaccessible. At this time, after data of the device 26 is migrated to, for example, the device 18, the data restoration control unit 8 sets the device 18 as a restoration target device and the device 1 as a log device.
9. If the data restoration unit 21 is started by specifying the log file 41 as the log file name, the time point 51 as the restoration start time, and the time point 53 as the end time, the data restoration unit 21 operates to Data at time 53
The data is updated.

(2) Restoration of data lost due to an operation error Assume that data of device 17 is migrated to archive device 26 at time point 51 and backup data is saved as in (1). It is also assumed that the log record of the device 17 has been collected in the log file 41. Time point 5
It is assumed that erroneous data is written to the device 17 due to an operation error between the time 2 and the time 53. At this time, after the data in the archive device 26 has been transferred to the device 17 and the data restoration unit 21 is started by designating the time point 51 as the restoration start time and the time point 52 as the end time, the data at the time point 53 on the device 17 is obtained. Can be returned to the data at the time point 52, so that the erroneous data between the time points 52 and 53 can be corrected by performing the operation again. If erroneous data related to only one log record due to an operation error, the data restoration unit 21 after data restoration
It is possible to restore the data of the device 17 to the latest state by referring to the log file 41 and using the subsequent log record.

(3) Reproduction of Past Data It is assumed that the data of the archive device 26 stores data up to the time point 51. It is also assumed that the log file 61 of the device 30 stores log records including the log records from the time point 51 to the time point 53. At this time, the data of the device 26 is migrated to, for example, the device 17, and the log file 61 of the device 30 is migrated to, for example, the device 18.
1. By specifying the time point 52 as the end time,
The data at the time point 52 can be reproduced.

It is also possible to apply the magnetic disk subsystem 2 as the storage subsystem 3, that is, as an archive subsystem, or to incorporate the data restoring unit 21 into the storage subsystem 3. In this case, since the storage subsystem 3 has the data restoration unit 21, the data restoration control unit 8 activates the data restoration unit 21 of the storage subsystem 3 so that the data of the archive device is restored from the restoration start time to the restoration end time. It can be updated by the archive log file 61. W
In step S6, the log file 61 after the completion of the update becomes unnecessary and is deleted. Even in the case where the storage subsystem 3 does not include the data restoration unit 21,
The WS 6 reads the log file 61 of the storage subsystem 3, reads and updates the data block of the corresponding archive device of the storage subsystem 3 by the log record, updates the data block of the archive device with the updated data block, In this way, the data in the archive device can be updated by the log file 61 until the target restoration end time. Also in this case, the log file 61 after the update is completed
May be deleted.

When the data of the archive device of the storage subsystem 3 is updated until the restoration end time as described above, the host 4 or 5 accesses the archive of the storage subsystem 3 instead of accessing the device of the magnetic disk subsystem 2. It becomes possible to access the device. That is, when the entire magnetic disk subsystem 2 becomes unusable due to a failure, the WS 6 is
And converts the address of the magnetic disk subsystem 2 and the number of the device to the corresponding address of the storage subsystem 3 and the number of the archive device by capturing the read / write command for the device of the magnetic disk subsystem 2 issued by It can be transmitted to the subsystem 3. The data of the archive device of the storage subsystem 3 is passed to the hosts 4 and 5 as the data of the corresponding device of the magnetic disk subsystem 2. W
In step S6, by accessing the monitoring information table 24 of the monitoring apparatus 1, the correspondence between the devices 11 to 19 of the magnetic disk subsystem 2 and the archive devices of the storage subsystem 3 having the backup data, and the restoration end time of each archive device Can be obtained.

FIG. 4 is a diagram showing the data structure of the LBA table 25. The LBA table 25 is created and accessed by the data restoration unit 21 in the second embodiment. Each record of the LBA table 25 has a device number, an LBA, and an update flag. The device number is the number of the device to be restored, LB
A is L sequentially assigned to the block in the device.
The BA and update flags are flags indicating whether or not the block has been added / updated. 0 indicates no addition / update, and 1 indicates addition / update. The records of the LBA table 25 are arranged in ascending order of LBA for the same device, starting from LBA0.

In the second embodiment, the data restoration unit 2
1 is activated by the data restoration control unit 8 and starts processing.
A BA table 25 is created, and all the update flags of each record are initialized to 0. Log file 4 specified next
1, the data of the device to be restored is updated by referring to the log record sequentially from the log record immediately before the restoration end time and in the direction of the restoration start time opposite to the time series. At this time, the start LBA and the number of blocks of each log record are referred to, and the LBA table 2 is used for the same LBA sequence.
If the update flag of the corresponding LBA in the LBA table 25 is 0 with reference to the corresponding LBA sequence of No. 5, the block is rewritten to 1 and the block data of the LBA on the restoration device is updated with the “contents” of the log record. If the update flag of the corresponding LBA in the LBA table 25 is 1,
The block data of the corresponding LBA on the restoration device is skipped. In other words, if the data is updated with the latest data for the block of the same LBA on the restoration device, it is unnecessary to update the data with the older data, so that the process is skipped. The processing of the data restoration unit 21 can be speeded up for an application in which data of the same LBA sequence is repeatedly updated by this processing.

For example, in the example of the log file 41 shown in FIG. 2, between the time point 51 and the time point 53, the blocks whose LBAs are 3 and 4 are updated three times. If the time point 53 is the restoration end time, it is sufficient to update the block with the LBA of 3 or 4 of the device by the log record 44 immediately before, and skip the update of the block with the same LBA of 3 or 4 before that. Can be.

The monitoring device 1 may be located far from the WS 6 via the LAN 33 and the Internet. Alternatively, the program and the monitoring information table 24 of the log collection control unit 7, the data restoration control unit 8, the data migration unit 9, and the status display unit 10 of the monitoring device 1 are built in the magnetic disk subsystem 2, and the display device 34 and the input device 35 May be connected to the magnetic disk subsystem 2. In this case, the log collection control unit 7, the data restoration control unit 8, and the data migration unit 9 of the magnetic disk subsystem 2 communicate with the WS 6 via the LAN 33.

[0040]

As described above, according to the present invention, log files generated in the storage subsystem have a common data format in the system, and log collection, data restoration, and data movement are performed. By executing the unit of device data accessed by a common logical device number as a unit, it is possible to provide a system for centrally managing backup data without regard to the format of data to be backed up. Also, since the data write time is stamped on each log record, if a device is associated with the log file from which the log was collected, the device data at a specific point in time is updated by the log record after that point by specifying the time. , Data can be restored. Therefore, data lost due to an operation error or the like can be restored. Also, by using log files corresponding to data in the archive device, data at a certain point in the past can be reproduced, and past data on a rewritable storage medium can be presented as evidence data.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a backup data management system according to an embodiment.

FIG. 2 is a diagram illustrating a data format of log files 41 and 61 according to the embodiment.

FIG. 3 is a flowchart illustrating a procedure of a data migration process according to the embodiment;

FIG. 4 is a diagram illustrating a data format of an LBA table according to the embodiment.

FIG. 5 is a diagram illustrating a conventional backup data creation situation.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Monitoring device, 2 ... Magnetic disk subsystem, 3 ...
-Storage subsystem, 4, 5-Host, 6-Workstation (WS), 7-Log collection control unit, 8-Data restoration control unit, 9-Data migration unit, 20 ... log collection unit, 21 ... data restoration unit, 41, 61 ... log file

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B018 GA04 HA04 KA03 MA12 5B065 CA12 CA15 CC08 CE22 CE23 EA12 EA33 EA35 5B082 DE06

Claims (3)

[Claims] 1. A first storage subsystem for storing data accessible from a host device on a rewritable storage medium, and a second storage subsystem for storing backup data of data of the first storage subsystem. A system for collecting a log of a data write time, a data storage location, and a data content requested by the higher-level device, and the data before update after the designated time. A first storage subsystem having means for updating and restoring data with a log, means for starting the means for collecting the log and means for restoring the data by specifying time, and when a specified time is reached. Means for relaying the movement of data between the first storage subsystem and the second storage subsystem; Backup data management system characterized by having a second storage subsystem to perform the data writing and data reading as the migration destination and migration source. 2. A first storage subsystem which is accessed from a higher-level device via a data transfer network,
And a second storage subsystem for storing backup data of the data of the storage subsystem, wherein the system has a log of a data write time, a data storage location, and a data content requested by the higher-level device. A first storage subsystem having means for collecting data, and means for updating data before update by the log after the designated time and restoring the data, and via the first storage subsystem and the control system network. A monitoring device that is connected and starts the means for collecting the log and the means for restoring the data at a specified time, and the monitoring device is connected to the monitoring device via the control system network and via the data transfer system network. Connected to a first storage subsystem and a second storage subsystem, and designated by the monitoring device; A computer which relays the movement of data between the first storage subsystem and the second storage subsystem when the time is reached; and data writing and data transfer destination and source of the data of the first storage subsystem A backup data management system, comprising: a second storage subsystem that performs reading. 3. A control means having an interface for communicating with an external device via a data transfer network, an interface for communicating with an external device via a control network, and a plurality of magnetic disk devices for storing data. Means for responding to a command via the control system network, collecting a log of a time of data writing to the magnetic disk device requested via the data transfer system network, a data storage location, and a data content. Means for updating data before update by the log after a designated time in response to a command via the control system network and restoring the data.