JP2001034540A - Difference backup method and backup restoring method in dually written data processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the deterioration of input/output performance and an unexpected increase in the quantity of acquired data and to make attainable a stable difference back-up operation by duplexing a control storage which stores the data updating identification information of a data storage and recording the identification information also in a control storage corresponding to a data storage that has not acquired yet the difference backup. SOLUTION: A data dualwrite control program 102 instructs a 1st controller 12 to update only a 1st data storage 14 for a prescribed time and also to record the 3rd identification information in a 3rd control storage 122. A difference backup program 103 instructs a 2nd controller 16 to record the 2nd identification information in a 2nd control storage 161 after all data of a 2nd data storage 18 are backed up and a collation character string is acquired in an external data storage 20. When the backup is over in a normal state, the OR of the 2nd and 3rd identification information is written in a 1st identification information storage area as the 1st identification information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a backup method, a backup restoration method, and a system for a data storage device accessed by a dual-write data processing system. Furthermore, even if the data storage device to be backed up fails, a method to minimize the effect on the system and reliably restore the data of the failed data storage device from the backup, and the execution time is short and the data acquisition amount is minimized To a method for allowing a differential backup to be continued.

[0002]

2. Description of the Related Art A data processing system that stores a large amount of data, such as a bank or an insurance company, obtains a large amount of backup in preparation for equipment failure or a catastrophic disaster. In order to reduce the storage cost of the backed up data, there is a demand for means for minimizing the amount of data to be backed up. In addition, data that has been backed up using recent high-speed, low-cost communication lines can be transferred to a server at a remote base and used for online operations the next day, or stored at a backup center in the event of a disaster. In order to effectively use communication lines and reduce communication costs, means for minimizing the amount of data transferred through the communication lines is required.

In response to the above demand, Japanese Patent Laid-Open Publication No. 57-9077
In the data processing system having a control storage device for storing identification information for identifying an area in a data storage device in which data has been updated, a backup of all data in the data storage device is temporarily stored in an external storage device. After the data is obtained by the device, a differential backup of only the data-updated area in the data storage device is obtained in the external storage device by referring to the identification information of the control storage device, and after the differential backup succeeds, the identification information is obtained. This example illustrates a method of erasing and preparing for the next differential backup, thereby shortening the backup execution time and minimizing the data acquisition amount.

[0004] On the other hand, due to recent business extension and continuous online operation for 24 hours, it is becoming difficult to temporarily suppress data update and secure time for executing backup. Therefore, there is a need for a means for acquiring a backup without affecting an application that involves updating data.

[0005] In response to the above-mentioned request
In Japanese Patent Application Laid-Open No. 5 (1999) -2005, in a data processing system in which a first data storage device connected to a central processing unit and a second data storage device at a remote place are connected by a high-speed link, update data from the central processing unit is received. The first data store creates a list of data to be copied to the second data store, without interference from the central processing unit,
By copying data from the first data storage device to the second data storage device and, if the copy is successful, updating the list, without affecting applications involving data updates running on the central processing unit. 4 illustrates a method of obtaining a backup in a second data storage device.

Therefore, in a conventional data processing system provided with a control storage device for storing information relating to data update disclosed in Japanese Patent Application Laid-Open No. 57-90770 in a second data storage device disclosed in Japanese Patent Application Laid-Open No. Hei 8-509565. Thus, a differential backup method has been realized in which the execution time is further reduced and the data acquisition amount is minimized without affecting applications involving data updates.

[0007]

In the conventional data processing system, when the second data storage device fails and the medium is replaced or replaced with a spare data storage device, the data of the second data storage device is replaced with the data of the second data storage device. The point at which the identification information of the control storage device that identifies the area in the updated second data storage device becomes inconsistent, or the time when the backup of the second data storage device was previously obtained when the control storage device failed. No consideration is given to the loss of the identification information for identifying the area in the second data storage device to which data has been updated thereafter, and all data in the first data storage device is copied to the second data storage device. Thus, the data of the second data storage device is restored while greatly affecting the input / output performance of the first data storage device, and all the data of the second data storage device are backed up. Data and data update 2 data storage device has a problem that must be overcome inconsistencies identification information in the control storage device for identifying a region of the second in the data storage device has been performed.

It is an object of the present invention to provide a first data storage device for storing identification information for identifying an area in a second data storage device or an area in the second data storage device in which data has been updated; By restoring the backup data to the second data storage device without affecting the input / output performance of the data storage device, and copying only the data not in the backup data from the first data storage device to the second data storage device, (2) The purpose is to minimize the influence on the input / output performance of the first data storage device when restoring the data in the data storage device, and further, without backing up all the data in the second data storage device. It is an object of the present invention to provide an improved method for enabling a differential backup of a second data storage device to be continued.

Another object of the present invention is to prevent a malfunction caused by using incorrect backup data when restoring the backup data to the second data storage device.

[0010]

To achieve the above object, a first control storage device for storing first identification information for identifying an area in a first data storage device in which data has been updated, A second control storage device for storing second identification information for identifying an area in the second data storage device that has been updated; a first data storage device that has been updated and a second data storage device that has not been updated; A third control storage device for storing third identification information for identifying an area in the first data storage device relating to the first data storage device, and a first data storage device instructed to update the first control device when the second data storage device can be updated. A data control unit instructs the second control device to update the area in the device also in the second data storage device, and instructs the third control storage device to record as the third identification information when the second data storage device cannot be updated. Overwriting system Function, the referring to third identification information 1
A copy function of deleting the third identification information after copying the updated data in the data storage device to the second data storage device;
A differential backup function is provided for erasing the second identification information after backing up to an external data storage device with reference to the second identification information.

First, the data duplication control function and the copy function are applied to match all data in the first data storage device and the second data storage device. Thereafter, the data update of the second data storage device is temporarily suppressed, and the update request during that time updates only the first data storage device and simultaneously records it as third identification information in the third control storage device. A backup of all data in the second data storage device is acquired in the external data storage device by applying the differential backup function, and recording of the second identification information is started. The data update inhibition of the second data storage device is released, the data duplication control function and the copy function are applied, and all data in the first data storage device and the second data storage device are made to match again. Updating of data in the data storage device is temporarily suppressed. When the backup of the data of the second data storage device is acquired in the external data storage device by applying the differential backup function, the data update of the second data storage device is temporarily performed by referring to the second identification information. After the suppression, the data update of the second data storage device is temporarily stopped until the update is temporarily stopped. Only the update data that has been updated in the first data storage device is selected, and the backup data is transferred from the second data storage device to the external data storage device. Can be obtained.

At this time, the identification information relating to the updated data obtained by updating only the first data storage device between the time when the data update of the second data storage device is temporarily inhibited and the time when the copy function is applied is the third identification information. Information is recorded. While the copy function is being applied to copy data from the first data storage device to the second data storage device, only the data stored in the first data storage device is updated and not copied to the second data storage device. Data that is being copied to the second data storage device after updating only one data storage device, data that has been copied to the second data storage device after updating only the first data storage device, The identification information on the data updated in both the first data storage device and the second data storage device is recorded in at least one of the second identification information and the third identification information. The updated data obtained by updating both the first data storage device and the second data storage device between the time when the copying by the copy function is completely completed and the time when the data update of the second data storage device is temporarily inhibited again is: It is recorded in the second identification information.

[0013] Therefore, the second identification information and the second identification information are stored between the time when the backup data was obtained from the second data storage device to the external data storage device and the time when the backup data was again obtained from the second data storage device to the external data storage device. By calculating the logical sum with the third identification information, the area in the first data storage device (that is, the first identification information) in which the data has been updated since the previously acquired backup data is obtained, and the first identification information is obtained.
Can be written to control storage.

By writing the logical sum of the second identification information and the third identification information as the first identification information in the first control storage device, the second data storage device fails and the medium is replaced or the spare storage device is replaced. When the data storage device is used as a substitute for the second data storage device, or when the second control storage device fails, when the data in the second data storage device is restored from the backup data in the external storage device, the data is updated after the backup data Can be obtained from the first identification information stored in the first control storage device. Therefore, the recording of the second identification information is started immediately after the second identification information is erased and the second control storage device starts recording from the state immediately after the backup of the second data storage device is acquired.
The identification information is written to the third control storage device as the third identification information. Then, by applying the copy function and referring to the third identification information, copying only the area in the first data storage device whose data has been updated after the backup data to the second data storage device, All data in the first data storage device and the second data storage device can be matched with minimum influence on the input / output performance of the storage device, and the second data updated after the backup data Since the area in the storage device is recorded in the second control storage device as the second identification information, the second control information is continuously stored in the second control storage device.
A differential backup of the data storage device can be obtained.

When restoring the data in the second data storage device from the backup data in the external storage device, if the wrong backup data is used, all the data in the first data storage device and the data in the second data storage device cannot be matched. . Therefore, a collation character string including information such as the number of times of acquiring the differential backup, the last backup acquisition date, and the volume serial number is added to the backup data, and this collation character string is also recorded in the first control storage device. When restoring the data in the second data storage device from the backup data, the collation character string of the backup data is collated with the collation character string of the first control storage device to verify whether the backup data is correct, and only when the backup data is correct. Erasing the second identification information, leaving the second control storage device in a state immediately after the backup of the second data storage device is obtained, and simultaneously writing the first identification information as the third identification information in the third control storage device. By doing so, it is possible to ensure that all data in the first data storage device and all data in the second data storage device coincide with each other without erroneous operation using erroneous backup data.

[0016]

FIG. 1, FIG. 2, FIG. 3, FIG. 4, FIG. 5, FIG. 6, FIG. 7, FIG.
0 and FIG.

First, an overall configuration diagram of an online data processing system according to an embodiment of the present invention will be described with reference to FIG. 1, and detailed configuration diagrams forming a part of FIG. 1 will be described with reference to FIGS.

FIG. 1 shows an embodiment of the present invention for 24 hours 3 hours.
It is a whole block diagram of a 65-day nonstop online data processing system. After creating a database to be referred and updated by the online program in the first data storage device and the second data storage device and starting the online program, every Sunday without interruption of the online program is performed at the time of midnight at midnight. A backup of all data in the two data storage devices is obtained from Monday to Saturday at 0:00 a.m., a differential backup of only the update data of the previous day's online program. The data double writing control function according to claim 1 is already started at the same time as the 24 hour 365 day non-stop online data processing system of FIG. 1 is started by the double writing control program 102 operating in the central processing unit. The third control storage device described in item 1 is the third control storage device 122 in the first control device 12, and the first data storage device described in claim 1 is 3436 cylinders from cylinder number 0 to cylinder number 3435. 2. The first data storage device 14 comprising 15 tracks from track number 0 to track number 14, the second data storage device according to claim 1 being 3436 cylinders from cylinder number 0 to cylinder number 3435, each cylinder being a track number. The second data storage device 18 is composed of 15 tracks from 0 to track number 14.

In FIG. 1, reference numeral 10 denotes a central processing unit for executing various programs; 101, an online program which operates in the central processing unit 10 to refer to or update the first data storage device 14 and the second data storage device 18; , 1
02 operates in the central processing unit 10 to update both the first data storage device 14 and the second data storage device 18 or only the first data storage device 14 and simultaneously update the data in the first data storage device 14. A double-write control program for instructing the third control storage device 122 to record the area as the third identification information.
0, the entire data of the second data storage device 18 or only the area in the second data storage device 18 indicated by the second identification information stored in the second control storage device 161 is stored in the external data storage device 20. The differential backup program 104 for backing up operates in the central processing unit 10 and
A copy program for copying all data in the data storage device 14 or data not updated in the second data storage device 18 with reference to the third control storage device 122 from the first data storage device 14 to the second data storage device 18; Reference numeral 105 denotes a backup restoration program which operates in the central processing unit 10 and restores all the data in the second data storage device 18 from the backup data in the second data storage device 18 backed up to the external data storage device 20.

Reference numeral 11 denotes a central processing unit 10 and a first control unit 1.
2 controls the input / output to / from the first data storage device 14 and the first control storage device 12
1 and the recording of the third identification information in the third control storage device 122.
The first control device, which responds from 02 to the reference request of the third identification information as to whether or not there is a record, 121 is the first data storage device 1
The first control storage device 122 for storing information relating to the data update of No. 4 is in the first control device 12, has been updated in the first data storage device 14, and has the second data storage device 1.
Reference numeral 8 denotes a third control storage device for storing information relating to unupdated data, 13 a bus connecting the first control device 12 and the first data storage device 14, and 14 a first control device 1
2 connected to cylinder number 0 to cylinder number 3435
Up to 3436 cylinders, each cylinder has track number 0
A first data storage device consisting of 15 tracks from track number 14 to track number 14 is a bus connecting the central processing unit 10 and the second control device 16.

Reference numeral 16 controls input / output to / from the second data storage device 18 and the second control storage device 161
A second controller 161 for controlling the recording of the identification information and responding to the request for reference to the second identification information from the differential backup program 103 whether or not the recording is present is in the second controller 16 and the second data storage device 18. A second control storage device for storing information relating to the data update of the second device; a bus 17 connecting the second control device 16 and the second data storage device 18;
Reference numeral 18 denotes 3436 cylinders which are connected to the second control device 16 and have cylinder numbers 0 to 3435, and each cylinder has 15 of track numbers 0 to 14
A second data storage device constituted by a track, 19 is a bus connecting the central processing unit 10 and an external data storage device 20, and 20 is connected to the central processing unit 10 and updates all data or updates in the second data storage device 18. An external data storage device for storing data.

FIG. 2 shows a configuration example of the first control storage device 121 of FIG. First identification information storage area 121
“a” stores, in the first data storage device 14, first identification information for identifying, in track units, an area in the first data storage device 14 where data has been updated. 1 cylinder 12
The first identification information 121b of the track is, of the first identification information stored in the first identification information storage area 121a, the m cylinder n track of the first data storage device 14 (1 cylinder 12 track in FIG. 2). Is 1-bit identification information indicating whether data has been updated by 0 (if not updated) or 1 (if updated). Collation character string storage area 1
Reference numeral 21c stores the collation character string added to the backup data when the backup was last obtained from the second data storage device 18 of FIG. 1 to the external data storage device 20.

FIG. 3 shows a configuration example of the second control storage device 161 of FIG. The second identification information 161a of one track and twelve tracks indicates whether or not data of m cylinder and n tracks (one cylinder and twelve tracks in FIG. 3) in the second data storage device 18 has been updated (0 (if not updated)). Or 1-bit identification information indicated by 1 (when updated).

FIG. 4 shows a configuration example of the third control storage device 122 of FIG. The third identification information 122a of one track and twelve tracks indicates whether or not the data of m tracks and n tracks (one cylinder and twelve tracks in FIG. 4) in the first data storage device 14 has been updated (if not updated). Or 1-bit identification information indicated by 1 (when updated).

Next, the operation will be generally described with reference to FIGS. 5, 6, and 7, and the details will be described with reference to FIGS. 8, 9, 10, and 11. FIG.

FIG. 5 is a schematic flowchart showing a method of obtaining a backup of all data in the second data storage device 18 at 0:00 am on Sunday in the external data storage device 20 in the system of FIG.

In step ST51, when the data in the first data storage device 14 and the data in the second data storage device 18 do not match when the online data processing system is started, the copy program 104 is started to store the first data. In a method of copying all data in the device 14 to the second data storage device 18, all data in the first data storage device 14 and all data in the second data storage device 18 are matched once.

At step ST52, the data duplication control program 102 suppresses the update of the second data storage device 18 at 0:00 am on Sunday, and thereafter, proceeds to step ST55.
Until the update inhibition of the second data storage device 18 is released, the first control device 12 updates only the first data storage device 14 and simultaneously records the third identification information 122a in the third control storage device 122. To instruct.

In step ST53, the differential backup program 103 is started to obtain a backup of all data in the second data storage device 18 and a collation character string in the external data storage device 20, and then control the second identification information 161a by the second control. The second control device 16 is instructed to record the data in the storage device 161.

In step ST54, step ST53
The second identification information 1
The logical sum of 61a and the third identification information 122a is written to the first identification information storage area 121a as the first identification information 121b. Further, the collation character string added to the backup data in step ST53 is written in the collation character string storage area 121c in the first control storage device 121.

In step ST55, in step ST52
In the second command specified by the data duplication control program 102
The update inhibition of the data storage device 18 is released.

In step ST56, the copy program 104 is started, and the third program started recording in step ST52.
The first data storage device 14 and the second data storage device 18 are copied by copying only the update data from step ST52 onward in the first data storage device 14 from the first data storage device 14 to the second data storage device 18 with reference to the identification information 122a. After matching all the data in the data storage device 18 again, the third identification information 1
The first controller 12 is instructed to stop recording at 22a, and prepares for online processing on Sunday and differential backup from Monday to Saturday.

FIG. 6 is a schematic flowchart showing a method of acquiring a backup of only updated data by the online processing of the previous day at 0:00 a.m. from Monday to Saturday in the system of FIG.

In step ST61, the data duplication control program 102 suppresses the update of the second data storage device 18 from 0:00 am on Monday to Saturday, and thereafter, the program proceeds to step ST61.
Until the update inhibition of the second data storage device 18 is released at T64, only the first data storage device 14 is updated, and at the same time, the first control device is configured to record the third identification information 122a in the third control storage device 122. Instruct 12.

In step ST62, the differential backup program 103 is started, and only the data updated by a request from the online program 101 after midnight of the previous day with reference to the collation character string and the second identification information 161a is read. 2 data storage device 18 to external data storage device 2
After acquiring the data to 0, the second identification information 161a indicating the data that has been backed up is all set to 0 (not updated), and the second control device 16 is instructed to start recording the second identification information 161a again. .

In step ST63, step ST62
When the differential backup program 103 has been completed normally, the logical sum of the second identification information 161a and the third identification information 122a is stored as the first identification information 121b in the first identification information storage area 121a. Further, the collation character string added to the backup data in step ST62 is written in the collation character string storage area 121c in the first control storage device 121.

In step ST64, step ST61 is executed.
In the second command specified by the data duplication control program 102
The update inhibition of the data storage device 18 is released.

In step ST65, the copy program 104 is started, and the third program started recording in step ST61.
The first data storage device 14 and the second data storage device 18 are copied by copying only the update data from step ST61 onward in the first data storage device 14 from the first data storage device 14 to the second data storage device 18 with reference to the identification information 122a. After matching all the data in the data storage device 18 again, the third identification information 1
The first controller 12 is instructed to stop recording at 22a, and prepares for online processing on the next day and backup at midnight on the next day.

FIG. 7 shows a failure of the bus 15, the second control device 16, the second control storage device 161, the bus 17, and the second data storage device 18 at any time after the step ST55 of FIG. 5 in the system of FIG. Occurs, and a new bus 15, a second control device 16, a second control storage device 161, a bus 17, and a second data storage device 18 are prepared by a method such as replacement of a storage device medium or replacement with a spare device. And FIG. 7 is a schematic flowchart until the operation of FIG. 6 returns to a state where it can be executed.

In step ST71, the backup restoration program 105 is activated to restore the backup data acquired in the external data storage device 20 to the newly prepared second data storage device 18.

In step ST72, the backup restoring program 105 checks the collation character string stored in the collation character string storage area 121c in the first control storage device 121,
Check against the check string added to the backup data. The process proceeds to the next step ST73 only when the two matching character strings match. If the two collation strings do not match, it is assumed that the wrong backup data has been restored, or that only part of the differential backup data obtained every day at midnight from Monday to the day of the failure has been restored, Steps ST71 and ST72 are repeated until the correct backup data is restored or all the differential backup data are restored.

In step ST73, the backup restoration program 105 sets all the second identification information 161a of the second control storage device 161 to 0 (not updated), and the differential backup program 103 transfers the backup of the second data storage device 18 to the external data storage. After the storage device 20 is in the same state as immediately after the acquisition, the second control device 16 is instructed to start recording the second identification information 161a again.

In step ST74, the backup restoration program 105 writes the first identification information 121b as the third identification information 122a in the third control storage device 122, and then records the third identification information 122a in the third control storage device 122. To the first control device 12 to perform the operation.

In step ST75, copy program 1
04, and refers to the third identification information 122a to copy only the data updated by the online program 101 after the backup data restored in step ST71 from the first data storage device 14 to the second data storage device 18. , All data in the first data storage device 14 and the second data storage device 18 are matched. After that, it instructs the first control device 12 to stop recording the third identification information 122a.

Next, the operations of FIGS. 5, 6 and 7 will be described in detail.

FIG. 8 is a detailed flowchart of steps ST51 and ST56 of FIG. 5, step ST65 of FIG. 6, and step ST75 of FIG.

In step ST81, it is determined whether all data in the first data storage device 14 and all data in the second data storage device 18 match. The first data storage device 14 and the second
If all data in the data storage device 18 match,
Since there is no need to activate the copy program 104 and make all data in the first data storage device 14 and the second data storage device 18 coincide with each other, steps ST82 and ST83 in FIG.
ST84, ST85, ST86 and ST87 are not executed.

In step ST82, the copy program 104 is started. The copy program 104 requests the first control device 12 that controls the recording of the third identification information 122a to refer to the third identification information 122a. First control device 1
When the recording of the third identification information 122a has not been started, the recording program 2 returns to the copy program 104 that the recording has not been started, and when the recording of the third identification information 122a has already been started, the recording has been started. I do.

In step ST83, copy program 1
04 is the first data storage device 14 and the second data storage device 1
In step ST8, a copy method for matching all data of step 8 is performed.
2 is selected based on the answer from the first control device 12.

In step ST84, step ST82
If the response from the first control device 12 has already started recording,
The copy program 104 refers to the third identification information 122a and copies only the update data of the first data storage device 14 to the second data storage device 18 by using a method of copying all of the first data storage device 14 and the second data storage device 18. Reconcile the data.

In step ST85, step ST82
When the answer from the first control device 12 is not recorded, the copy program 104 copies the entire data of the first data storage device 14 to the second data storage device 18 by using the first data storage device 14 and the second data storage device 18. All data in the data storage device 18 is made to match.

In step ST86, it is determined whether the copy executed by the copy program 104 in step ST84 or step ST85 ends normally or abnormally. In the case of a normal end, the process proceeds to the next step ST87, and in the case of an abnormal end, the copying is restarted from the step ST82.

In step ST87, step ST86
When the copy program 104 ends normally and all the data in the first data storage device 14 and all the data in the second data storage device 18 match, the copy program 104
The first controller 12 is instructed to stop the recording at 22a.

Step ST84 or step ST85
When an update request is received from the online program 101 while the copy program 104 is copying data from the first data storage device 14 to the second data storage device 18, the copied area in the first data storage device 14 In the case of an update request for the first data storage device 14 and the second data storage device 18, both are updated. In the case of an update request for the pre-copy execution area in the first data storage device 14, only the first data storage device 14 is updated. Is updated, and in the case of an update request for the copy-in-progress area in the first data storage device 14, the data dual-write control program 102 is instructed to wait for the update until the copy is completed. Japanese Patent Publication No. 2544449 discloses a method of making all data in two data storage devices coincide with each other while performing a copy operation and data update from a program in parallel in the dual-write data processing system.

FIG. 9 is a detailed flowchart of step ST52 of FIG. 5 and step ST61 of FIG.

In step ST91, the data duplication control program 102 executes the copy program 104 for updating the second data storage device 18 and the backup restoration program 105 before inhibiting the update of the second data storage device 18 in step ST93. Investigate whether you are running
If neither the copy program 104 nor the backup restoration program 105 has been executed, the next step S
After executing T92, if at least one of the copy program 104 and the backup restoration program 105 is being executed, the process waits until the process ends.

In step ST92, the data duplication control program 102 suspends the update request from the online program 101 until step ST95, and the data in the second data storage device 18 whose update is inhibited in step ST93 and in step ST94. Third identification information 1 to start recording
22a is ensured.

In step ST93, the data duplication control program 102 suppresses data update of the second data storage device 18 until step ST95, and an update request from the online program 101 updates only the first data storage device. To do it.

In step ST94, the data duplication control program 102 instructs the first control device 12 to start recording the third identification information 122a in the third control storage device 122.

In step ST95, the data duplication control program 102 releases the update request from the online program 101 held in step ST92 and executes it. However, in step ST93, the update request from the online program 101 updates only the first data storage device.

FIG. 10 shows steps ST53 of FIG. 5, FIG.
5 is a detailed flowchart of step ST62 of FIG.

In step ST101, the differential backup program 103 is started.

In step ST102, the differential backup program 103 writes a collation character string including the number of times of backup acquisition, the last backup acquisition date, and the volume serial number to the external data storage device 20 prior to the backup acquisition in step ST105 or step ST106.

In step ST103, the differential backup program 103 requests the second control device 16 for controlling the recording of the second identification information 161a to refer to the second identification information 161a. The second control device 16 controls the second identification information 161a.
If the recording of the second identification information 161a has not been started, the recording of the second identification information 161a has been started.
Answer 03.

In step ST104, the differential backup program 103 executes a method for backing up data from the second data storage device 18 to the external data storage device 20 in step S104.
The selection is made based on the answer from the second control device 16 in T103.

In step ST105, step ST1
03, the recording from the second control device 16 has already been started, the differential backup program 103 refers to the second identification information 161a and writes only the updated data in the second data storage device 18 in step ST102. The data is added to the column and backed up to the external data storage device 20.

In step ST106, step ST1
03 when the answer from the second control device 16 is unrecorded,
The differential backup program 103 backs up the external data storage device 20 by adding all data in the second data storage device 18 to the collation character string written in step ST102.

In step ST107, it is determined whether the backup executed by the differential backup program 103 in step ST105 or ST106 has been completed normally or abnormally. In the case of normal termination, the process proceeds to the next step ST108, and in the case of abnormal termination, the process proceeds to step ST108.
Perform backup again from 102.

In step ST108, when the differential backup program 103 ends normally, all the second identification information 161a is set to 0 (not updated), and all data in the second data storage device 18 is backed up.
It instructs the second control device 16 to start or restart the recording of the second identification information 161a, and proceeds to the next step.

FIG. 11 shows steps ST54 of FIG. 5, FIG.
6 is a detailed flowchart of step ST63 of FIG.

In step ST111, the differential backup program 103 stores the second identification information 161a in the second control device 16 which controls the recording of the second identification information 161a in the step ST53 in FIG. 5 or the step ST62 in FIG. Request a reference. The second control device 16 stores the second identification information 161a in the differential backup program 103.
Answer.

In step ST112, the differential backup program 103 stores the third identification information 122a in the first control device 12 which controls the recording of the third identification information 122a in the step ST52 in FIG. 5 or in the step ST61 in FIG. Request a reference. The first control device 14 stores the third identification information 122a in the differential backup program 103.
Answer.

At step ST113, the differential backup program 103 sets the second identification information 161a (0 if not updated, 1 if updated) referenced at step ST111 and the third identification information 1 referenced at step ST112.
The logical sum of the first identification information storage area 12 in the first control storage device 121 is calculated by calculating the logical sum of the first identification information 121b and the logical sum of the first identification information storage area 121b and the result of the calculation.
The process of writing to 1a is executed for all cylinder numbers and all track numbers.

In step ST114, the differential backup program 103 stores the collation character string determined in step ST102 of FIG. 10 and written in the external data storage device 20 in the collation character string storage area 121 in the first control storage device 121.
Also write to c.

According to the present embodiment, the second identification information 161a
The logical sum of the first identification information 122a and the third identification information 122a
1b in the first identification information storage area 121a, the bus 15, the second control device 16, and the second
Control storage device 161, bus 17, second data storage device 1
8, the first data storage device 14
Storage device 20 without affecting the input / output performance of the
The data in the second data storage device 18 is restored from the acquired backup data, and only the area in the first data storage device 14 whose data has been updated since the backup data is stored in the first data storage device 14 using the first identification information 121b. By copying the data from the device 14 to the second data storage device 18, the influence on the input / output performance of the first data storage device 14 is minimized and all the data in the first data storage device 14 and the second data storage device 18 are copied. There is an effect that can be matched. Further, the second data storage device 18 whose data has been updated after the backup data by the copy.
Since the middle area is recorded in the second control storage device 161 as the second identification information 161a, there is an effect that the differential backup of the second data storage device 18 can be continuously obtained.

Further, according to this embodiment, the collation character string of the backup data is collated with the collation character string in the collation character string storage area 121c, and the above processing is performed only when the backup data is correct. There is an effect that all data in the first data storage device 14 and the second data storage device 18 can be surely matched without malfunction using data.

[0077]

As described above, the control storage device for storing the identification information relating to the data update of the data storage device is duplicated, and the control storage device corresponding to the data storage device for which the differential backup has not been obtained is also provided. By recording the identification information, even if the data storage device for which the differential backup was acquired or the corresponding control storage device fails, the impact on the I / O performance of the system is minimized and the system is restored and continued at the same time As a result, the differential backup can be performed, so that it is possible to prevent the input / output performance from deteriorating due to the recovery work and to prevent an unintended increase in the amount of acquired data, thereby achieving a stable differential backup operation.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of an online data processing system according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of a first control storage device 121 in FIG. 1;

FIG. 3 is a configuration diagram of a second control storage device 161 in FIG. 1;

FIG. 4 is a configuration diagram of a third control storage device 122 in FIG. 1;

FIG. 5 shows a second data storage device 1 in the system of FIG.
8 is a schematic flowchart showing a method of acquiring a backup of all data of No. 8;

FIG. 6 is a schematic flowchart showing a method for acquiring a backup of only update data by online processing of the previous day in the system of FIG. 1;

FIG. 7 shows a second data storage device 1 in the system of FIG.
FIG. 7 is a schematic flowchart showing a process until the operation of FIG.

FIG. 8 shows steps ST51 and ST56 in FIG.
7 is a detailed flowchart of step ST65 of FIG. 6 and step ST75 of FIG.

9 is a step ST52 of FIG. 5 and a step ST of FIG.
FIG.

10 is a flowchart showing step ST53 in FIG. 5 and step S in FIG. 6;
It is a detailed flowchart of T62.

11 is a step ST54 of FIG. 5 and a step S of FIG. 6;
It is a detailed flowchart of T63.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... Central processing unit, 102 ... Double writing control program, 103 ... Differential backup program, 104 ... Copy program, 105 ... Backup restoration program, 12 ... First control device, 121 ... First control storage device
121a: first identification information storage area, 121b: first identification information, 121c: collation character string storage area, 122: third control storage device, 122a: third identification information, 14: first data storage device, 16: first 2 control device, 161: second control storage device, 161a: second identification information, 18: second data storage device, 20: external data storage device.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kazuya Higuchi 5030 Totsukacho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Software Division of Hitachi, Ltd. (72) Norio Oikawa 5030 Totsukacho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Hitachi, Ltd. Software Division (72) Inventor Masashi Kondo 5030 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture F-term (Reference) 5B018 GA04 HA04 MA12 QA01 5B082 DA02 DC06 DE06 GA04

Claims (4)

[Claims] 1. A first data storage device, a first control device for controlling input / output of the first data storage device, and first identification information for identifying an area in the first data storage device in which data has been updated. A first control storage device for storing data, a second data storage device, a second control device for controlling input / output of the second data storage device, and an area in the second data storage device where data has been updated. A second control storage device for storing second identification information, wherein the first data storage device is updated and the second data storage device identifies an area in the first data storage device for unupdated data. A third control storage device for storing third identification information, a central processing unit for instructing the first control device and the second control device to input / output data, and a first control device when the second data storage device can be updated.
The second control device is instructed to update the area in the first data storage device to which the control device is instructed to update, and if the second data storage device cannot be updated, the third identification information is used as the third identification information. A data duplication control function for instructing the control storage device to record, a copy for deleting the third identification information after copying the updated data of the first data storage device to the second data storage device with reference to the third identification information A dual write data processing system comprising: a differential backup function of deleting the second identification information after backing up the external data storage device with reference to the second identification information,
From the start of backup of the update data from the second data storage device to the external data storage device with reference to the second identification information until the completion of the next backup of the update data from the second data storage device to the external data storage device A differential backup method, further comprising means for creating first identification information from the second identification information and the third identification information and writing the first identification information to the first control storage device. 2. The system according to claim 1, wherein the second
From the time when the backup of the update data from the second data storage device to the external data storage device is started with reference to the identification information to the time when the backup of the update data from the second data storage device to the external data storage device is completed next time, 2 The collation character string to be added to the backup data of the data storage
A backup method, further comprising a means for writing to a control storage device. 3. The system according to claim 1, wherein when restoring the backup data of the second data storage device acquired in the external storage device to the second data storage device, the second identification information is deleted and the first identification information is deleted. A means for creating third identification information from a third device and writing it to a third control storage device. 4. In the system according to claim 1, when the backup data created by the backup method according to claim 2 is restored to the second data storage device by the backup restoration method according to claim 3, the collation character of the backup data. 4. A backup restoring method, characterized in that restoration success and restoration failure are determined based on the result of collating the collation string with the collation character string in the first control storage device, and the means according to claim 3 is executed only when the restoration is successful.