JP2003076593A - Database management method and system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform a referring/updating processing based on contents of a database (DB) at a certain point of time in the past while ensuring successive access by a normal job to the DB. SOLUTION: The database (DB) 8 is divided into a first storage area 9, a second storage area 10 and a third storage area 11 and managed. In this case, in the first storage area 9, latest data at the present point of time by a normal application (UAP) operated under an online system are stored. Also, in the second storage area 10, data (t0 data) to be frozen at a certain point (t0 ) of time in the past are stored. In the third storage area 11, the data (tn data) after updating which are a result of executing a new UAP are stored. A DB access control part 5 receives specification of latest or a certain point of time in the past from the UAP 4 and controls reference/updating of the respective storage areas 9, 10 and 11 of the DB 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management technique for storing a database at an arbitrary time point in another area and processing the database.

[0002]

2. Description of the Related Art Generally, daily or monthly processing of a database (hereinafter referred to as a DB) of an online system is performed after the completion of online or after the DB is blocked and temporarily disconnected from the online system. This is done by collectively referring to the DB in processing.

Conventionally, as a DB updating / searching method in a continuous non-stop system, as described in JP-A-58-62744 and JP-A-6-35774, a DB is updated.
While the target information is immediately updated to the update request, the pre-update information is retained and the update time-series information is added to it, and the search request corresponds to the time-series information at the time of the request. There is a way to make it possible to retrieve time information.
Further, as a method for facilitating the recovery of the DB and the generation management of the records, as described in JP-A-1-166232, the record update time and the pre-update information are built in the DB as a history. There is a method to realize it. Further, as a method of acquiring a backup of a data set, as described in Japanese Patent Laid-Open No. 5-210555, a backup of a data set having a consistency is acquired at a certain time while significantly reducing the hold time of UAP execution. There is a way to do it. However, the purpose of each of these is to read out the consistent contents of the DB at a certain point in time, and cannot be directly updated.

[0004]

In the prior art, while guaranteeing continuous access to the DB by the original work, a separate update process or the like is performed based on the contents of the DB at some point in the past (hereinafter referred to as frozen data). In order to do, DB at a certain point
You will make a copy of and then update it. In addition, when developing UAP under the online system environment, D
The work of recovering B from the backup was required each time prior to the execution of UAP.

It is an object of the present invention to easily enable update processing of daily work and monthly work in a database management system.

[0006]

In order to achieve the above object, the present invention provides a database management method in a database management system having a first storage area, a second storage area and a third storage area as follows. To achieve.

First, a processing request by the execution of the first program is input, database processing is performed based on the processing request, and data based on the processing result of the database processing is stored in the first storage area as a database. . Next, when the predetermined condition is satisfied, the database stored in the first storage area is stored in the second storage area. That is, the database stored in the first storage area is duplicated in the second storage area. A second program that performs processing is executed by using the database stored in the second storage area as an input, and update data for the processing is stored in the third storage area.

By performing the processing as described above, the update data of the processing with the second storage area storing the database when the predetermined condition is satisfied can be stored in the third storage area, so that the continuous operation is performed. The update processing such as daily work and monthly work in the database management system can be easily realized.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION An example of an embodiment of the present invention will be described. First, a DB is divided into a plurality of storage areas and managed.
Specifically, the first storage area that holds all the latest data by the UAP processing route (normal processing route) of the normal operation of the online system is set as the center axis, and a certain point in time (t ₀ A second storage area for storing data before update (t ₀ data) for tracing back to a point in time) and a processing route of another UAP (another processing route) from a certain point in the past based on the t ₀ data Is executed and the updated data (t _n data) is divided into the third storage areas for management.

Next, if it is not necessary to go back to a certain point in the past, it is only necessary to separate the second storage area and the third storage area. When setting a new time point, the initialized second storage area and third storage area may be connected in time series or in parallel. These operations do not prevent the normal business access to the first storage area of the DB. Therefore, continuous access to the DB is guaranteed.

Corresponding to the division of the DB into a plurality of storage areas, the DB access control section receives a designation of the latest or a certain point in the past from the UAP and searches a predetermined storage area for a reference request. Means for storing the updated data in the first storage area or the third storage area at the time of updating, and the means for storing the data in the second storage area before updating at the time of storing in the first storage area. Means for saving data are used in combination.

An embodiment of the present invention will be described below with reference to the drawings.
To do. <System environment> FIG. 1 shows a database management system (DBM of this embodiment).
The system environment in which S) operates is shown. In Figure 1, middle
Central processing unit (CPU) 1, terminal group 6, system console
Online system with database 7 and database 8
Are configured. Here, the CPU1 is operated
System (OS) 2, transaction processing monitor
(TP monitor) 3, user application program
(UAP) 4 and DB access control unit 5 are included.
OS2 is software that operates the entire system efficiently.
Is. The TP monitor 3 receives the traffic input from the terminal group 6.
Transaction is received via OS2 and the desired business process is performed.
To UAP4 that executes UAP4 is DB access
The control unit 5 is called to perform data input / output processing. D
The B access control unit 5 responds to the reference or update request from the UAP4.
It is a control unit that accesses the DB 8 according to
It is a central part of the process. DB8 is this online
The original UAP process (normal process)
1st memory that stores the latest data of the route)
Rear 9 and some time in the past of the normal processing route (t₀)
Data (t₀Second storage area 10 for storing data)
And the above t ₀New UAP processing based on the data
Data that has been updated by executing the processing (another processing route).
(T_nConsists of a third storage area 11 for storing data)
To be done. The second and third storage areas 10, 11 are implemented
It is a characteristic configuration in the example, each storage area
Assign one or more datasets to 10, 11
Although it is possible to use
I am going to each one.

From the system console 7, a system operator inputs a command to the DB access control unit 5, that is, a command for dynamically connecting / disconnecting data sets corresponding to so-called storage areas 9, 10, 11. FIG. 1 shows the connected state. <Configuration of UAP and DB Access Control Unit> Requests for accessing the DB 8 from the UAP 4 to the DB access control unit 5 are roughly classified into a reference system and an update system. Specifically, when the unit of access is a record, the reference system refers to record reference, and the update system refers to record replacement, record addition, and record deletion. Here, the replacement / addition / deletion requests are generically called update requests. FIG. 2 shows a configuration example of the UAP 4 and the DB access control unit 5.

The UAP 4 includes a reference request means 21 and an update request means 22. Further, the reference request means 21 has the latest data 211, t ₀ data 212, and t _n data 2
An item for selecting any one of 13 is included. The update request means 22 has the latest data 221 and t _n data 222.
Includes an item to select one of. Incidentally, there is no item for selecting t ₀ data in the update request unit 22, a t _n data that is obtained as a result of updating the t ₀ data, t ₀
The data update request is meant to mean a t _n data update request.

The DB access control unit 5 includes the latest data search means 231, t ₀ data search means 232, t _n data search means 233, and the latest data storage means 241.
Each storage means of the t ₀ data storage means 242 and the t _n data storage means 243 and the first storage area read control means 25
1, second storage area read control means 252, third storage area read control means 253, first storage area write control means 261, second storage area write control means 262, and third storage area write control means 263 of each writing means. The solid line shows the control route at the time of reference request, and the broken line shows the control route at the time of update request. The operation for each data reference request or update request will be described later. <Time point setting> There are two methods of setting the time points of t ₀ and t _n from the UAP, that is, an event method and a schedule method.

The event-based method means that the time when the setting request is issued from the UAP itself becomes the setting time.
Interpreted as the time to start storing t ₀ data or t _n data.

The schedule method is a method of designating a system time prior to the time when the setting request is issued from the UAP. In other words, it means the reservation reception from UAP, and when the system reaches the designated time after reception,
The MS automatically starts storing t ₀ or t _n data. <Processing Route and Update of Record> FIG. 3A shows a basic tree of a normal processing route and another processing route. After the initial creation of the DB at the time of creating the DB 31, the normal processing route (shown by the solid line)
Is a time t ₀ of 32 based on the DB creation time 31.
Has been set, and it has reached 33 at the present time. This route represents the latest data. On the other hand, another processing route (indicated by a dashed line) after the time t ₀ of 32 is set, 34 time t _n as a base to said time point t ₀ is set, leading to the present time 35. This processing route represents t _n data.

FIG. 3 (2) shows each processing route.
One record (record A) is being updated
Indicates a child. Of the 36 records created at the time of DB creation 31
The content is A, and then the normal processing route update 37
Becomes B, and at time t 32 ₀Update 38 after setting
Therefore, it becomes C. Therefore, the latest data is C,
t₀The data is B. On the other hand, t₀The data B is 34
Time t_nBy another process route from 39 by D
Becomes Therefore, t_nThe data is D.

FIG. 3 (3) shows the state after that. That is, D becomes E by the update 391 in another processing route from the time point t _n at 34. On the other hand, C of the latest data of the normal processing route becomes F by the update 381. As a result, the latest data at the current times 331 and 351 is F, the t ₀ data is B, and the t _n data is E.

Plural time points t ₀ and t _n may be set in time series on each processing route. FIG. 4 (1) shows an example of the normal processing root tree set twice at the time point t ₀ . That is, the normal processing route is 41 at the time of DB creation.
After that, the setting 42 at the first time point t ₀ [1], the setting 43 at the second time point t ₀ [2], and the current time point 44 are reached. Further, FIG. 4 (2) shows an example of another processing root tree set twice for the time point t _n . That is,
This is because, after the DB creation time point 45 and the time point t ₀ setting 46, another processing route sets the time point t _n [1] setting 47 and the time point t _n [1] for the first time based on the time point t ₀ . The setting 48 of the second time point t _n [2], which is the base, and the current time point 49 are reached. It should be _{noted that} in both cases of the time points t ₀ and t _n , the setting of three times or more can be easily analogized.

Although FIG. 4 shows an example in which a plurality of time points t ₀ and t _n are set in time series on the same processing route, a plurality of different processing routes can be set based on the time point t ₀ .

FIG. 5 shows an example in which two different processing routes are set with two times t _n set on the basis of the same time t ₀ . That is, after the DB creation time 51 and the time t ₀ setting 52, the time t ₀ 5 is set separately from the normal route.
There is a first alternative processing route that sets 53 the first time t _n 1 based on 2 and reaches the present time 54. On the other hand, the second time point t _n 2 based on the same time point t ₀ 52 is set 55.
However, there is a second alternative processing route up to the present time 56. These two processing routes are common at the base time point t _0, but form independent processing routes after the time point t _n . <Operation of DB Access Control Unit> FIGS. 6 to 11 are flowcharts of processes performed by each unit of the DB access control unit 5 of FIG.

FIG. 6 is a flow chart of the latest data searching means 231 which operates in response to the latest data reference request. When the latest data reference request 211 is issued from the UAP 4, the latest data search means 231 calls the first storage area read control means 251 (step 2311), D
The first storage area 9 of B8 is accessed to obtain the latest target data. After that, the latest data search means 231 returns the control to the UAP 4.

FIG. 7 is a flowchart of the latest data storage means 241 which operates in response to the latest data update request. When the latest data update request 221 is issued from the UAP 4, the latest data storage unit 241 determines whether the time point t ₀ is set (step 2411), and if not set, the first storage area write control unit 261 is immediately executed.
Is called (step 2413), and the target latest data is stored in the first storage area 9 of the DB 8. On the other hand, t ₀
If is set, first, the data before update is t ₀
In order to store the data in the second storage area 10 as data, the t ₀ data storage means 242 is called (step 241).
2) After the execution of the t ₀ data storage means 242, the latest data storage means 241 calls the first storage area write control means 261. Then, when the processing of the update request of the latest data is completed, the latest data storage unit 241 returns the control to the UAP4.

FIG. 8 corresponds to the update request of the latest data.
To work₀Detailed flow chart of the data storage means 242
It is a chart. Called from the latest data storage means 241
Was t₀The data storage means 242 first stores the second data of the DB8.
Corresponds to memory area 10₀Whether the data is stored
The second storage area read control means 252 for confirmation
(Step 2421), the second storage area 10
To access. In addition, the second storage area 10 is a time series
Row t₀[1] to t₀When divided into [n], the last
At time t₀The area [n] is targeted. That
Applicable t₀It is determined whether there is data (step 242).
2) If there is, the latest data storage means 241 without doing anything
Return control to. On the other hand, applicable t₀If there is no data,
The first storage area read control means 251 is called (scan
Step 2423), access the first storage area 9
It That is, if there is corresponding data in the first storage area 9,
This is t ₀It becomes data. Next, write to the second storage area
Call the control means 262 (step 2424), first
The pre-update data read from the storage area 9 is t₀data
As a result, it is stored in the second storage area 10. in this case
Also, the second storage area 10 is time-series t₀[1] to t₀[N]
When the area is divided into
Be targeted. In addition, the record addition to the first storage area
In this case, t is stored in the first storage area.₀No data exists
In the second storage area, an empty t₀data
(No data part 72 of FIG. 12 described later) is written. That
After t₀The data storage means 242 is the latest data storage means 2
Return control to 41.

[0026] FIG 9 is a flowchart of t ₀ data retrieval means 232 which operates in response to a reference request for t ₀ data. When there is a reference request 212 for t ₀ data from the UAP 4, the t ₀ data searching means 232 first calls the second storage area read control means 252 (step 232).
1), the second storage area 10 of the DB 8 is accessed to determine whether there is corresponding data (step 232).
2). If there is corresponding data in the second storage area 10, it is the t ₀ data to be obtained. t ₀ data search means 232
Passes the t ₀ data read from the second storage area 10 to the UAP 4 and returns the control to the UAP 4. The second
Reference memory area 10 be divided into a time series _{t 0 [1] ~t 0 [} n], when t ₀ [1] is the time set in order from the specifying the time t ₀ [i] In response to the request, the corresponding data is t ₀ [i] → t ₀ [i + 1] → ... t ₀
The search is performed in the order of [n], and the search is terminated at the first found.

When there is no target t ₀ data in the second storage area 10, the t ₀ data retrieval means 232 calls the first storage area read control means 251 (step 232).
3) Access the first storage area 10 of DB8. That is, if there is corresponding data in the first storage area 10,
That is the required t ₀ data. Therefore, the t ₀ data searching means 232 obtains the data read from the first storage area 9 as t ₀ data and returns the control to the UAP 4.

FIG. 10 is a flow chart of the t _n data storage means 233 which operates in response to a t _n data reference request. When a reference request 213 of t _n data from UAP4, t _n data retrieval means 233 first calls the third storage area read control means 253 (Step 23
31), the third storage area 11 of the DB 8 is accessed to determine whether there is the corresponding t _n data (step 23).
32). If there is corresponding data in the third storage area 11, it is the t _n data obtained. t _n data search means 233
Passes the t _n data read from the third storage area 11 to the UAP 4 and returns the control to the UAP 4. The third
If the storage area 11 is to be divided into a time series _{t n [1] ~t n [} n], is set to n from 1 in order, t _n
Only [1], t _n [2] ... t _n [n-1] are referred to, but t _n [n] is an area to be referred to and updated. Now, _if a reference request specifying t _n [i + 1] is accepted, t _n [i] → t _n [i−1] → ... t
_n [2] → to find the appropriate data in the order of t _n [1], the search ends at the first one found. If the latest t _n data is requested, the search is started from the area of t _n [n].

If the target t _n data does not exist in the third storage area 11, the t _n data searching means 233 calls the t ₀ data searching means 232 (step 2333), and as described in FIG. Alternatively, the first storage area 9 is further accessed. That is, if there is t ₀ data corresponding to the second or first storage area, it is the t _n data to be obtained. The t _n data search means 233 passes the t _n data to the UAP 4, and returns control to the UAP 4.

FIG. 11 is a flow chart of the t _n data storage means 243 which operates in response to a t _n data update request. When the t _n data update request 222 is issued from the UAP 4, the t _n data storage means 243 calls the third storage area write control means 263 (step 243).
1) The target t _n data is stored in the third storage area 11 of DB8. Note that the third storage area 11 is time-series t
_n [1] have been divided into ~t _{n [n],} if it is set up to n than 1 in _{order, t n [1], t} n [2] ··· t n
Only [n-1] is referred to, but _tn [n] is an area where reference and update are performed. Therefore, t _n data is written in the area of t _n [n].

The operation of the DB access control unit 5 will be described below with reference to the examples of (2) and (3) of FIG. 3 with reference to the flowcharts of FIGS. 6 to 11.

It is assumed that the record A has already been stored in the first storage area 9 by the creation 36 at the DB creation time point 31. Since the record B of the update 37 has not yet been set at the time point t ₀ , it is immediately transferred to the first storage area write control means 261 by the latest data storage means 241 and stored in the first storage area 9. That is, the data in the first storage area 9 is updated from A to B.

After the setting 32 at the time point t ₀ , the operation of the DB access control unit with respect to the reference request and the update request is as follows. In response to the latest data reference request, the latest data search means 231 causes the first storage area read control means 251 to operate, and the first storage area 9 to record B
Is obtained (FIG. 6: step 2311). In the update 38, the latest data storage means 24 responds to the update request of the record C.
The t ₀ data storage means 242 is called by 1 (FIG. 7: step 2412). t ₀ data storage means 24
2 calls the second storage area read control means 252, but issues a notification that the pre-update data is not stored in the second storage area 10 to the second storage area read control means 252.
Further, (FIG. 8: Steps 2421 and 2422), the first storage area read control means 251 is subsequently called, and the record B which is the pre-update data is received from the first storage area read control means 251 (FIG. 8: Step 2423), second storage area writing control means 26
2 (FIG. 8: Step 2424). As a result, the record B is stored in the second storage area 10 as t ₀ data. On the other hand, the updated data record C is passed to the first storage area write control means 261 by the latest data storage means 241 (FIG. 7: step 241).
3), the record C is stored in the first storage area 9 as update data.

Next, it is assumed that there is a setting 34 at the time point t _n and a reference request for t _n data is generated. To a reference request of the t _n data, t _n data retrieval means 233, first to the third
The storage area read control means 253 is called, but the third
A notification that the corresponding data is not stored in the storage area 11 is received from the third storage area read control means 253 (FIG. 10: steps 2331 and 3332), and then the t ₀ data search means 232 is called (FIG. 10: step 23).
33). The t ₀ data retrieval means 232 calls the second storage area read control means 252, and as a result, receives the record B stored as t ₀ data in the second storage area 10 from the second storage area read control means 252. (FIG. 9: Steps 2321 and 2322 (Ye
s)), and passes this to the t _n data search means 233. Thus, the record B is obtained. Then, update 39 of t _n data
Occurs, the record D of the update request is passed to the third storage area write control means 263 called by the t _n data storage means 243 (FIG. 11: step 243).
1), stored in the third storage area 11.

Subsequently, reference and update 39 of t _n data are performed.
Suppose 1 occurs. First, the request for reference t _n data, t _n data retrieval means 233 calls the third storage area read control unit 253, a third storage area 11
It is received from the third storage area read control means 253 that the record D of the corresponding data is stored in (FIG. 10: Steps 2331 and 3332 (Yes)). In this case, the target record D is obtained only by reading the third storage area 11. After that, the record E of the update request of the update 391 is passed to the third storage area write control means 263 called by the t _n data storage means 243 (FIG. 11: step 2431) and stored in the third storage area 11.
As a result, E is replaced with D in the third storage area 11, and the first and second storage areas 9 and 10 are stored.
, C and B are stored in the memory.

Finally, in the update 381 of the latest data, the latest data storage means 241 calls the t ₀ data storage means 242 (FIG. 7: step 2412), and the t ₀ data storage means 242 uses the second storage area read control means. 252
To receive a notification that the pre-update data is stored in the second storage area 10 (FIG. 8: step 24).
21, 2422 (Yes)) and the call of the second storage area write control means 262 is bypassed. This point is different from the operation of the update 38 described above. After that, the latest data storage unit 241 calls the first storage area write control unit 261 (FIG. 7: step 2413). As a result, the record F of the update request of this time is stored in the first storage area 9, and the record B remains stored in the second storage area 10. <Data Format of Second and Third Storage Areas> The first storage area 9 can be applied to a DB of a generally known hierarchical model, network model, or relational model. The data storage format can be realized by the conventional DB data format itself. However, the second and third storage areas 10 and 11 require a data format different from that of the first storage area 9 because time-series area management should be performed for update-type record access. That is, it is a record relating to addition / deletion of a record.

As a concrete example, FIG. 12 shows the data formats of the second and third storage areas 10 and 11. The header part 71 is composed of a flag field 711, a key length field 712, a record length field 713 and a key field 714, and the data part 72 is composed of only a record field 721. Flag field 7
In 11, an update flag, an addition flag, and a deletion flag are set as access types. The key length field 712 represents the size of the key field 714, and the record length field 713 represents the size of the record field 721. When a record is added to the first storage area, the value of the record length field is zero, that is, information without a data part is stored as t ₀ data in the second storage area. As described in the memory of t ₀ data of FIG. 8, it is necessary t ₀ data storage in step 2424, even if in step 2423 the record addition there is no corresponding data in the first storage area, without data portion t ₀ data Write out. When the record of the t ₀ data is deleted, the information in which the value of the record length field is zero is also stored as the t _n data. In step 2431 of FIG. 11, t _n data without a data part is written out.

As described above, the t ₀ data in the second storage area and the t _n data in the third storage area are stored in the key field 7.
14 logically associates with the data in the first storage area.

The access performance of the record can also be improved by cooperating with each other. For this, a buffering method or a parallel execution control method by association can be considered. In addition, for the second or third storage area, the header section 71 is used as necessary.
There is a method that has an index by the key of. Further, in the case of the second and third storage areas, if the data is divided into n in time series, the search efficiency (also the space storage efficiency) can be improved by merging continuous divided areas.

As described above, in the database management system, since the DB is composed of a plurality of storage areas and the data for each purpose is centrally managed, it is possible to copy the data to another file and store the duplicated data as in the conventional case. No need. Therefore, data management becomes easy. Furthermore, since there is no duplication of data, DB storage space is efficient, and D
There is an effect that the time required for copying B is saved.
On the other hand, in terms of business and operation of the system, the contents of the DB at a plurality of points in the past can be collectively referred to later, which facilitates statistical processing and backup processing. In addition, while guaranteeing continuous access to the DB, update processing that starts from a certain point in the past can be multiplexed in multiple storage areas, so DB space is also available for UAP development (test work, etc.) in an online environment. The amount and the number of reloads (re-creation) can be deleted, and the DB can be used efficiently.

[0041]

According to the present invention, by having the third storage area, the update processing in the database management system during continuous operation can be easily realized.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a system in which an online database management system of the present invention operates.

FIG. 2 is a diagram showing a configuration of an embodiment of a DB access control unit according to the present invention and a diagram showing a relation with a UAP.

FIG. 3 is a diagram showing an example of a processing root tree and how data is updated according to the present invention.

FIG. 4 is a diagram showing an example in which a plurality of time points are set on a processing route.

FIG. 5 is a diagram showing an example in which a plurality of independent processing routes are set.

FIG. 6 is a processing flowchart of a latest data reference request of a DB access control unit.

FIG. 7 is a processing flowchart of a latest data update request of a DB access control unit.

FIG. 8 is a processing flowchart of t ₀ data storage of the DB access control unit.

FIG. 9 is a processing flowchart of a t ₀ data reference request of the DB access control unit.

FIG. 10 is a processing flowchart of a t _n data reference request of the DB access control unit.

FIG. 11 is a processing flowchart of a t _n data update request of the DB access control unit.

FIG. 12 is a diagram showing an example of a data format of second and third storage areas.

[Explanation of symbols]

1 Central processing unit (CPU) 2 Operating system (OS) 3 Transaction processing monitor (TP monitor) 4 User Application Program (UAP) 5 Database access control unit 8 Database (DB) 9 First storage area 10 Second storage area 11 Third storage area

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Masahiro Funagawa             6-81 Onoe-cho, Naka-ku, Yokohama-shi, Kanagawa             Hitachi Software Engineering Stock Association             In-house (72) Inventor Masao Yamagishi             6-81 Onoe-cho, Naka-ku, Yokohama-shi, Kanagawa             Hitachi Software Engineering Stock Association             In-house (72) Inventor Sumio Watanabe             5030 Totsuka Town, Totsuka Ward, Yokohama City, Kanagawa Prefecture             Ceremony Company Hitachi Software Development Division F term (reference) 5B075 UU40                 5B082 FA16

Claims (11)

[Claims] 1. A database management method in a database management system having a first storage area, a second storage area and a third storage area, wherein a processing request by execution of a first program is input, and the processing request is input to the processing request. A database processing based on the processing result of the database processing is stored as a database in the first storage area, and when a predetermined condition is satisfied, the database stored in the first storage area is stored in the second storage area. Stored in the second storage area, executes a second program for performing processing by using the database stored in the second storage area as an input, and stores updated data of the processing in the third storage area. Database management method. 2. The database management method according to claim 1, wherein the predetermined condition is a predetermined time, and the first database at the predetermined time is stored in a second storage area. 3. The predetermined condition is an input of a predetermined request, and the first database at the time when the predetermined request is issued is stored in a second storage area.
Database management method described. 4. An area for storing a database stored in a first storage area at a plurality of times is allocated and stored in the second storage area, and the second program corresponding to the plurality of areas is stored. The database database management method according to claim 1, wherein each of the data is executed, and the updated data by the processing of the second program is allocated and stored in each area of the third storage area. 5. An area for storing update data by processing of a plurality of the second programs, which inputs a database stored in one second storage area, is associated with the second program, and a third area is stored. The database management method according to claim 1, wherein the data is allocated to a storage area and stored. 6. A database management system having a first storage area, a second storage area and a third storage area, wherein a processing request by execution of a first program is input and database processing based on the processing request is input. Means for storing data based on the processing result of the database processing as a database in the first storage area, and a database stored in the first storage area when the predetermined condition is satisfied An area storing means, and a means for executing a second program for performing processing by using the database stored in the second storage area as an input and storing update data of the processing in the third storage area. A database management system characterized in that 7. The predetermined condition is a predetermined time, the second condition
7. The database management system according to claim 6, wherein the storage means stores the first database at the predetermined time in the second storage area. 8. The predetermined condition is an input of a predetermined request, and the means for storing in the second storage area stores the first database at the time when the predetermined request is issued in the second storage area. 7. The database management system according to claim 6, wherein: 9. A means for allocating and storing an area for storing a database stored in a first storage area at a plurality of times in the second storage area, and a second means corresponding to the plurality of areas. 7. The database database management system according to claim 6, further comprising means for executing each program and allocating and storing the update data by the processing of the second program in the area of the third storage area. 10. The means for storing in the third storage area comprises:
An area for storing update data by the processing of the plurality of second programs, which inputs a database stored in one second storage area, is allocated to the third storage area in association with the second program. The database management system according to claim 1, which is stored. 11. A database management system having a first storage area and a second storage area, inputting a processing request by execution of a first program, performing database processing based on the processing request, and performing the database processing. Means for storing data based on the processing result of the above in the first storage area as a database, and means for storing the database stored in the first storage area at a predetermined time in the database processing in the second storage area A database management system comprising: