JP2003006010A - Method and device for controlling data transition, and processing program of the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data transition controller to transit data having a network structure data base to data having a relational data base. SOLUTION: The data transition controller 100 is provided with a data type transition processing part 102 to transit data base definition 112 of data on the basis of the network structure data base system 110 (DB hereafter) to table definition 122 of a relational data base system 120 (RCB hereafter) and a record transition processing part 104 to transit a mater record A114-1 and a slave record B114-2 in parent-child relationship with the master record of the network hierarchy DB 114 to the same line of the table of an RDB 124.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data migration technique between systems having different database bases, and more particularly, to a data migration control device for migrating network structure database base data to relational database base data, its method and its processing. Regarding the program.

[0002]

2. Description of the Related Art In recent years, a relational database (hereinafter referred to as RDB) has become mainstream as a database (hereinafter referred to as DB) used in a computer system, but a conventional network structure database (hereinafter referred to as network structure DB). The system built with
Many still exist today. The network structure DB is
This is a system in which a pointer is used for a data record to define and associate a parent-child relationship between a parent record and a child record, and records according to a hierarchical tree structure.

The users who use this network structure DB have a very strong demand for system migration to an RDB, which facilitates data operation and system operation. A DB system migration method is disclosed in, for example, Japanese Patent Laid-Open No. 10-307742 (hereinafter referred to as Document 1). In this document 1, when each record of the network structure DB is migrated to the RDB, a relation key is added by referring to a pointer, so that the parent record and the child record before the migration are each table having a relationship. Have each remembered.

[0004]

[Problem to be Solved by the Invention] Network structure DB
In, the parent-child relationship of records is managed by pointers, and the information of related child records is narrowed down when the parent record is accessed. Therefore, the I / O generated when accessing the child record is relatively small.

However, when the migration is performed by the method described in the above-mentioned Document 1, the application after migration has access to search the relationship between the table corresponding to the parent record and the table corresponding to the child record. Since the instruction is issued, the number of I / Os increases. As a result, the processing capacity is lowered. In order to solve this problem, the data of network structure DB base is changed to RDB.
It is necessary to review and change the data structure when moving to the base. The change in the data structure greatly affects the entire processing logic of the application that processes the data. For this reason, the amount of renovation accompanying the change has increased, and in proportion to this, the burden of work time and work cost for migrating the DB system has increased.

The present invention has been made in view of the above problems, and an object thereof is data for migrating network structure DB-based data to RDB-based data in a short time without degrading the performance after application migration. It is to provide a migration control device, its method, and its processing program.

[0007]

(1) In order to achieve the above object, a data migration control device according to the present invention is a network structure in which a parent-child relationship is constructed between a parent record and a child record by a pointer. In a data migration control device for migrating database-based data to RDB-based data, a parent record is migrated to an RDB table, and correspondingly, a child record having a parent-child relationship with this parent record is converted into a table parent. It is characterized by having a record transfer means for transferring a record to the same line as the line to which the record has been transferred.

(2) Further, in the data migration control device according to the present invention, in the above (1), for each of the parent record and the child record referred to based on the parent-child relationship, a plurality of data types constituting the record are created. An abstract data type migration means for defining a corresponding abstract data type and for generating a user-defined function for performing data operation in a relational database with respect to a parent record and a child record in which the abstract data type is defined. Is characterized by.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to FIGS. FIG. 1 is a block diagram showing a configuration example of the entire DB system including a data migration control device in the first exemplary embodiment. FIG. 2 is a flow chart for outlining the operation of the data type migration processing unit in the first embodiment. FIG. 3 is a diagram for explaining the operation of the data type migration processing unit according to the first embodiment in more detail. FIG. 4 is a flowchart showing an outline of the operation of the record migration processing unit in the first embodiment. FIG. 5 is a diagram for explaining the operation of the record migration processing unit according to the first embodiment in more detail. FIG. 6 is a block diagram showing a configuration example of the entire DB system including the data migration control device in the second exemplary embodiment. FIG. 7 is a flowchart for outlining the operation of the data operation instruction change processing unit provided in the data migration control device according to the second embodiment. FIG. 8 is a diagram for explaining the operation of the data operation instruction change processing unit in the second embodiment in more detail. FIG. 9 is a diagram for explaining a record search processing example of the application converted by the data operation instruction change processing unit according to the second embodiment.

(First Embodiment) FIG. 1 shows an overall view of a DB migration system in which a data migration control device 100 according to the present invention is used. This DB migration system includes a data migration control device 100 and a network hierarchy DB.
It has a system 110 and an RDB system 120. The data migration control device 100 is connected to the network hierarchy DB system 110 and the RDB system 120.

The network hierarchy DB system 110 is
It has a DB definition 112 and a network hierarchy DB 114. As described above, the network hierarchy DB system 110 is a system that uses a pointer to a data record to define and associate a parent-child relationship between a parent record and a child record, and records according to a hierarchical tree structure. This parent-child relationship is defined in the DB definition 112. The network hierarchy DB 114 stores various data based on the parent-child relationship. RDB system 120
Does not have a hierarchical structure in the table itself, and the relationship with other tables is mutually defined through specific column data in the table.

The data migration control device 100 comprises a data type migration processing unit 102 and a record migration processing unit 104. The data type migration processing unit 102 refers to the data type of the DB definition 112 of the network hierarchy DB system 110, and sets the table definition 1 to the RDB system 120.
A DB definition having the same data type as 22 is generated.
The record migration processing unit 104 causes the RDB 124 to store various data stored in the network layer DB 114 of the network layer DB system 110.

The operation of the data migration control device 100 shown in FIG. 1 will be described below with reference to FIGS.

First, the operation of the data type migration processing unit 102 shown in FIG. 1 will be described with reference to FIGS. FIG. 2 shows an outline of the operation of the data type migration processing unit 102. The data type migration processing unit 102 uses the network layer D.
Parent-child set AB of DB definition 112 in which related information of record type A 112-1 and record type B 112-2 is stored, which is stored as DB definition of B system 110
From (not shown), the record type definitions of the record type A 112-1 and the record type B 112-2 are acquired. The data type migration processing unit 102 uses the record type A112-1 as an abstract data type A122-1 as an abstract data type and the record type B112-2 as an abstract data type B122-2 as an abstract data type B122-2. The data types are created in association with each other (step S201). A user-defined function of record A and a user-defined function of record B, which are functions defined for performing data operations of the abstract data type A 122-1 and the abstract data type B 122-2, are created (step S202).

In addition, it is searched whether or not the record type to be migrated is stored as a DB definition (step S).
203). If there are no other record types to be migrated,
It is searched whether or not there is a parent-child set type to be transferred (step S204). If there is no parent-child set type to be migrated,
A table definition defining table information in the RDB is created based on the created abstract data type (step S207). If there is a parent-child set type to be migrated, an abstract data type corresponding to the parent-child set type is created (step S205). Furthermore, it is searched whether there is another parent-child set type to be migrated (step S20).
6). If there is a parent-child set type to be migrated, again create an abstract data type corresponding to the parent-child set type in step S205.
Then, the same processing is performed thereafter.

Next, with reference to FIG. 3, the data type migration processing unit 102 described in FIG. 2 (the data type migration processing unit 3 in FIG. 3 will be described.
(Corresponding to 00) will be described in detail. As the DB definition, a record type A312 (corresponding to the record type A112-1 in FIG. 1), a record type B314 (record type B112 in FIG. 1).
-2), and parent-child relationship information 316 (not particularly shown in FIG. 1) are defined.

Based on the record type A312, the data type migration processing unit 300 defines the abstract data type definition 322a of the record A and the user-defined function definition 32 of the record A.
2b is generated. The data type of the abstract data type definition 322a is the same as that of the record A312 before the migration. The user-defined function 322b of the record A is a function that enables manipulation / execution of SQL (structured query language) data.

Similarly, the data type migration processing unit 300 generates the abstract data type definition 324a of the record B and the user-defined function definition 324b of the record B based on the record type B314. The data type of the abstract data type definition 324a is the same as that of the record B314 before migration. The user-defined function 324b of the record B is a function that enables manipulation / execution of SQL data.

Further, the data type migration processing section 300 generates an abstract data type definition 326a named record AB and a definition 326b named record AB based on the parent-child set AB definition 316. Abstract data type definition 3
26a defines a record type AB. This record type AB is an abstract data type record A of the definition 322a.
And the column of the repeating structure of the record B of the abstract data type of the definition 324a.

Here, the column of the repeating structure refers to a column of the array structure in which the number of elements of the array is variable, and is a kind of the column of the multi-valued attribute. By making the column of record B a column having a repeating structure, the parent-child relationship is expressed in the same record. The table definition 326b is a record A of an abstract data type called the record type AB declared in the definition 326a.
It is composed of a column named B and defines the logical structure of the table of record AB.

Next, the operation of the record migration processing unit 104 shown in FIG. 1 will be described with reference to FIGS.

FIG. 4 shows an outline of the operation of the record migration processing section 104 in FIG. The record migration processing unit 104 searches the parent record (here, the record A114-1) in the network hierarchy DB 114 (step S401). The record migration processing unit 104 determines whether or not a parent record exists (step S402).
If the parent record exists, the record migration processing unit 104 is the corresponding record, which is the network hierarchy DB 1.
The 14 child records (here, record B114-2) are searched (step S403). Record migration processing unit 1
04 determines whether a child record exists (step S404). If it exists, the record migration processing unit 104 assigns the parent record and the child record to the corresponding RDB.
It is stored in line 124 (step S405). If there is no child record in step 404, only the parent record is stored in the corresponding row of the RDB 124 (step S406).

Next, the operation of the record migration processing unit 104 of FIG. 1 (corresponding to the record migration processing unit 500 in FIG. 5) will be described in detail with reference to FIG. In the network hierarchy DB system 510, it is a parent record, and
The parent record 531 having the content “A1” and the content “A1”
2 ”532 and the parent record 532. In addition, the content is “B1” which is a child record of the parent record 531.
A child record 533 having a parent record 532 and a child record 534 having a content “B2” which is a child record of the parent record 532.
Child record 535 having the content “B3”. The parent record A114-1 in FIG. 1 is replaced with the parent records 531 and 532 in FIG. 4 and the child record A1 in FIG.
14-2 corresponds to the child records 533, 534 and 535 of FIG. 4, respectively.

The record migration processing unit 500 refers to the parent record 531 and the child record 533 and stores them in the row 551 in the RDB 520. The parent record 531 and the column data 541 have the same content “A1”. Child record 533
And the column data 543 have the same content “B1”. The relationship that the parent record 531 and the child record 533 have a parent-child relationship is related to the same line 551 in the RDB system 520.
Stored and identified in.

Similarly, the record migration processing section 500 refers to the parent record 532 and the child records 534 and 535 and stores them in the row 552 in the RDB 520. The parent record 532 and the column data 542 have the same content “A2”. The child record 534 and the column data 544 have the same content “B2”. The child record 535 and the column data 545 have the same content “B3”. The relationship that the parent record 531 and the child records 534 and 535 have a parent-child relationship is stored and identified in the same row 552 in the RDB system 520.

A column having a repeating structure can have a plurality of data in the same row. Therefore, column data 5
Both 44 and 545 are stored in the same row 552. In this way, by storing records having a parent-child relationship in the network hierarchy DB 510 in the same row of the RDB 520, it becomes possible to identify the relationship.

In the data migration control device according to the present invention, I
By performing a process of migrating from network structure DB-based data to RDB-based data so as to create a data structure that suppresses an increase in the number of / Os, system migration processing is performed without degrading data processing performance. be able to.

(Second Embodiment) FIG. 6 shows an overall view of a DB migration system in which a data migration control device 600 according to the present invention is used. This DB migration system includes a data migration control device 600, a network hierarchy DB
It has a system 610 and an RDB system 620. In the present embodiment, the data migration control device 600 further includes a data operation command change processing unit 606 as compared with FIG. Further, in the present embodiment, it is further clarified that the application is stored in the network hierarchy DB system 610 and the RDB system 620, as compared with FIG. The data migration control device 600 includes a network hierarchy DB system 610.
And the RDB system 620.

The operation of the data operation command change processing unit 606 which performs the characteristic processing of the data migration control device 600 shown in FIG. 6 will be described below with reference to FIGS. 7 to 9.

FIG. 7 is a flow chart for explaining the outline of the operation of the data operation command change processing unit 606. First, the data operation instruction change processing unit 606 retrieves the data operation instruction stored in the application storage unit 616 (step S701).

The data operation command change processing unit 606 determines whether or not the data operation command is a search command (step S7).
02), if it is a search command, it is further determined whether or not the operation target record is a parent record (step S703).
If it is a parent record, it is converted into a search SQL for a column having an abstract data type corresponding to the parent record (step S
704). If the result of determination in step S703 is not a parent record, it is converted to search SQL for a column having an abstract data type corresponding to the child record (step S70).
5).

Next, it is determined whether or not the data operation instruction is an update instruction (step S706), and if it is an update instruction, it is further determined whether or not the operation target record is a parent record (step S707). If it is a parent record, update SQL for a column having an abstract data type corresponding to the parent record
(Step S708). If it is not the parent record, it is converted into the update SQL for the column having the abstract data type corresponding to the child record (step S709).

Next, it is determined whether or not the data operation instruction is a storage instruction (step S710).
Further, it is determined whether or not the operation target record is a parent record (step S711). If it is a parent record, the storage SQ for the column having the abstract data type corresponding to the parent record
Convert to L (step S712). If it is not the parent record, it is converted into the update SQL for the column having the abstract data type corresponding to the child record (step S713).

Next, it is judged whether or not the data operation instruction is a deletion instruction (step S714).
Further, it is determined whether or not the operation target record is a parent record (step S715). If it is a parent record, delete SQ for the column having the abstract data type corresponding to the parent record
Convert to L (step S716). If it is not the parent record, it is converted into the update SQL for the column having the abstract data type corresponding to the child record (step S717). Finally, it is determined whether or not all the data operation commands have been searched (step S718), and if there is any remaining, the process returns to the first step S701 to continue the processing, and if there is no remaining, the process ends.

Next, with reference to FIG. 8, an example of a record search and an example of a record update will be described regarding the modification of an existing application.

FIG. 8 shows an example of data access processing at the time of record retrieval of the application processing section changed by the data operation instruction change processing section 800 (corresponding to the operation instruction change processing section 606 of FIG. 6) of the present invention. . In the search example in the network hierarchy DB system 810 (corresponding to the network hierarchy DB system 610 in FIG. 6) in FIG.
System 820 (corresponding to RDB system 620 in FIG. 6)
Is an example of a search in In the search example in the network hierarchy DB system 810,

[0037] (A1) FIND FIRST record A WHERE item AA =? ? GET record A (B1) FIND NEXT record B IN parent-child set AB GET record B (C1) FIND NEXT record B IN parent-child set AB GET record B

It is assumed that the data operation instruction is written. In (a1), the FI specifying the search condition for the item AA for the parent record “record A”
Perform a record search with the ND sentence, and find the record that hit
Get by ET sentence. The row 852 of the RDB system 820 will be described below as an example. Note that "A1", "B1", etc. indicate the contents of the record. In this example, the record 832 having the content “A2” is acquired. (B1)
Then, it is a search in the parent-child set for the child record “record B”. The FIND sentence is used for retrieval, and the hit record is obtained by the GET sentence. In this example, the record 834 of “B2”, which is a child record of the parent record of “A2”
Has been acquired. (C1) is a search within the parent-child set for the child record similar to (b1), and the record 835 of the child record "B3" is acquired.

When such data operation is changed by the data operation change program 800, it becomes as follows.

[0040] (A2) DECLARE cursor A CURSOR   FOR SELECT record AB. Record A     FROM record AB       WHERE         Record AB. Record A. Item AA =? ? OPEN cursor A FETCH cursor A INTO: PA (B2) DECLARE cursor B CURSOR   FOR SELECT Y. CB     FROM Record AB AS X       UNNEST (X. record AB. Record B) AS Y (CB)       WHERE         X. Record AB. Record A. Item AA =? ? OPEN cursor B FETCH cursor B INTO: PB (C2) FETCH cursor B INTO: PB

(A2) RDB the data operation of (a1)
It was converted for. The FIND statement in the network hierarchy DB corresponds to the SELECT statement in RDB,
In addition, DECLARE CU is a method that uses a cursor to handle search results for RDB one by one.
Replace with RSOR statement. DECLARE CURSO
In the description of the search in the R sentence, the record name “record A” to be searched in (a1) is set to “record AB” as the table name to be searched in (a2), and the column name to be searched is “record AB”. Record AB. Record A ". Furthermore, the search condition expression is (a1) and "item AA =?
? The column name is added in (a2) and replaced with “record AB. Record A. item AA = ??”. OP after DECLARE CURSOR statement
Describe the EN sentence to enable searching using the cursor. G which is an instruction to receive the search result in (a1)
The ET sentence is replaced with the FETCH sentence in (a2). As a result, in (a2), the column data 842 of "A2", which is the same value as the search result in (a1), can be received.

(B2) RDB the data operation of (b1)
It was converted for. In the RDB, the child record search in the network hierarchy DB is replaced with a search for a column having a repeating structure corresponding to the child record part. At this time, the UNNEST function of the SQL statement is used so that each element of the column of the repeating structure can be treated like a row of one table. The UNNEST function is a function added in the international standard SQL99, and is a function of referring to a column portion of an array structure in a table and referring to array elements like a row in the table.

On the basis of this idea, as in the case of (a2), the FIND statement in the network hierarchy DB is made to correspond to the SELECT statement in RDB, and the cursor is used to handle the search results for RDB one by one. Is replaced with the DECLARE CURSOR statement. In the description of the search in the DECLARE CURSOR statement, the alias "X" of the table name "record AB" is declared and used for convenience of description, and the column portion "X. record AB. Record B" of the repeating structure is used. UNNE
It is also called "Y (CB)" by looking at it by the ST function.
Is declared and used.

In the search of (b1), the search condition is not specified, but this is a search within the parent-child set, and the parent record and the child record of the network hierarchy DB correspond to the same row of one table in the RDB. In addition, the search condition of “record AB. Record B = ??” of (a2) corresponding to the search condition for the parent record is added as a search condition for identifying the row for the data operation of (b2). . After the DECLARE CURSOR statement, OPEN
Write a sentence so that you can search using the cursor. GET which is an instruction to receive the search result in (b1)
The sentence is replaced with the FETCH sentence in (b2). As a result, in (b2), the column data 844 of "B2" having the same value as the search result of (b1) can be received.

(C2) RDB is the data operation of (c1)
It was converted for. The FIND sentence of (c1) is (b
Since it is the next child record search after the search in 1) and the cursor declared in (b2) can be used as it is in RDB, it is not replaced with another instruction in RDB. The GET statement that is an instruction to receive the search result in (c1) is F in (c2).
Replace with ETCH sentence. Accordingly, in (c2), the column data 845 of "B3", which is the same value as the search result in (c1), can be received. Note that the RDB system 820 can also be applied by applying the above-described operation to the records 831 and 833 of the network hierarchy DB system 810.
Line 851 of FIG.

FIG. 9 shows an example of processing when a record of an application program changed by the data operation instruction change processing unit 900 (corresponding to the data operation instruction change processing unit 606 of FIG. 6) of the present invention is updated. Network hierarchy DB system 910 (network hierarchy DB system 6 of FIG. 6)
(Corresponding to 10) and the RDB system 920
(Corresponding to the RDB system 620 in FIG. 6).

In the update example in the network hierarchy DB system 910, (d1) STORE record A (e1) STORE record B

The data operation command is described.
In (d1), a record is added to the parent record “record A”. Here, the RDB system 92
Row 953 in 0 will be described below. In addition,
“A1”, “B1”, etc. indicate the contents of the record. In this example, a record 936 of “A3” is added. In (e1), "Record B", which is a child record
Record is added to. Record 9 in this example
A record 937 of “B4” is added as a child record of 36. When such a data operation is changed by the data operation change program 900, it becomes as follows.

[0049] (D2) INSERT INTO record AB (record AB. Record A) VAL UES (Record A (: PAA,: PAB) (E2) UPDATE record AB   ADD record AB. Record B (*) = Record B (: PBA,: PBB)     WHERE record AB. Record A. Item AA =: PAA

The data operation (d2) is a conversion of the data operation (d1) for RDB. The STORE statement for the parent record in the network structure DB is RD
In B, it is replaced with the INSERT statement. The record name “record A” to be updated in (d1) is replaced with “record AB” as the table name to be updated and “record AB. Record A” as the column name to be updated in (d2). The value to be stored is specified by VALUES, but the column to be stored is an abstract data type column, so the user-defined function “record A (: PAA,: PAB)” is stored in VALUES.
(PAA and PAB are parameter names) is specified. (D2)
The column data 946 having the content of "A3" is stored by the data operation of.

The data operation of (e2) is a conversion of the data operation of (e1) for RDB. The STORE statement for the child record in the network structure DB is RD
In B, replacement is performed by adding an element of the column of the repeating structure in the UPDATE statement. UPDATE for the RDB system
It is assumed that the function of adding the element of the column of the repeating structure is mounted by the ADD designation of the sentence. In (e1), the record name “record B” to be updated is set to “record AB” as the table name to be updated in (e2), and “Record AB. Record B” as the column name to be updated.
(*) ”. The value to be stored is specified by ADD, but since the column to be stored is an abstract data type column, the user-defined function “record B (: PBA,: P
BB) "(PBA and PBB are parameter names) are designated.

Furthermore, a WHERE specification which is a condition for specifying the line to be the object of UPDATE is added. U
Since the target row of PDATE is the row added in (d2), the table "record A" of the values stored in (d2)
By specifying the value corresponding to the primary key "B" in the WHERE condition, the row to be updated is specified. In the example of FIG. 5, it is assumed that the primary key of the table “record AB” is “record AB. Record A. item AA”, and the WHERE condition is “record AB. Record A. item AA =: PAA” (where PAA is a parameter name). ) Is specified. The column data 947 having the content of "B4" is stored by the data operation of (e2). In addition,
Record 93 of network hierarchy DB system 910
By applying the above-mentioned operation to 1,932 and the child record corresponding thereto, the line 95 of the RDB system 820
1, 952 respectively.

As described above, by performing the data operation command change processing, the application can be updated at the same time as the data migration, and the migration work time can be shortened.

The present invention is not limited to the examples described with reference to FIGS. 1 to 9 described above, and various modifications can be made without departing from the scope of the invention. For example, in the first embodiment, a plurality of data structures can be accepted using the abstract data type, but the accepted data type may be set in advance.

In the first and second embodiments, the number of child records is 1 or 2 for the sake of convenience in the network hierarchy DB system. However, it is natural that three or more child records are set. Applicable.

Further, in the above-described embodiment and the above-described various modifications, the program for performing the processing may be stored in a recording medium such as a hard disk as application software. This makes it possible to store the program or the like in a portable recording medium such as a CD-ROM and sell or carry it. It should be noted that the programs described in FIGS. 8 and 9 are merely examples, and the programs are not limited to these.

[0057]

According to the data migration control device, the method thereof, and the program thereof according to the present invention, the network structure DB-based data is changed to the RDB-based data so as to create the data structure for suppressing the increase of the I / O frequency. By performing the process of migrating to, the system migration process can be performed in a short time without degrading the data processing performance.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration example of an entire DB system including a data migration control device according to a first embodiment.

FIG. 2 is a flowchart showing an outline of an operation of a data type migration processing unit in the first embodiment.

FIG. 3 is a diagram for explaining the operation of a data type migration processing unit in the first embodiment in more detail.

FIG. 4 is a flowchart showing an outline of an operation of a record migration processing unit in the first embodiment.

FIG. 5 is a diagram for explaining the operation of a record migration processing unit in the first embodiment in more detail.

FIG. 6 is a block diagram showing a configuration example of an entire DB system including a data migration control device according to a second exemplary embodiment.

FIG. 7 is a flowchart showing an outline of an operation of a data operation command change processing unit provided in a data migration control device in the second embodiment.

FIG. 8 is a diagram for explaining the operation of the data operation instruction change processing unit in the second embodiment in more detail.

FIG. 9 is a diagram for explaining a record search processing example of an application converted by a data operation instruction change processing unit in the second embodiment.

[Explanation of symbols]

100, 600 ,: data migration control device 102, 300, 602 ,: data type migration processing unit 104, 500, 604 ,: record migration processing unit 110, 310, 510, 610, 810910: network hierarchy DB system 112, 612: DB definition 114, 614: Network hierarchy DB 120, 320, 520, 620, 820, 920: R
DB systems 122, 622: table definitions 124, 624, RDB 606, 800, 900: data operation command change processing units 616, 626: application storage unit

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Yuji Toyama             6-81 Onoe-cho, Naka-ku, Yokohama-shi, Kanagawa             Hitachi Software Engineering Stock Association             In-house F-term (reference) 5B082 CA10

Claims (5)

[Claims] 1. A data migration control device for migrating network structure database-based data, in which a parent-child relationship is constructed between a parent record and a child record by a pointer, to relational database-based data, wherein the parent record is a relational one. Record transfer means for transferring to a database table and correspondingly transferring the child record having the parent-child relationship with the parent record to the same row as the row to which the parent record of the table has been transferred; A data migration control device characterized by the above. 2. The data migration control device according to claim 1, wherein, for each of the parent record and the child record referred to based on the parent-child relationship, abstract data corresponding to a plurality of data types forming a record. An abstract data type migration means for defining a type and generating a user-defined function for performing data operation in the relational database with respect to the parent record and the child record in which the abstract data type is defined. Data migration control device. 3. The data migration control device according to claim 1, wherein a data processing command created for the network structure database and executing an application execution process is associated with the record migration means. The data migration control device further comprising a data operation command changing means for changing the data processing command for the above table. 4. A data migration control method for migrating network structure database-based data in which a parent-child relationship between a parent record and a child record is constructed by a pointer to a relational database-based data, wherein the parent record is a relational database table. And correspondingly, the child record having the parent-child relationship with the parent record is moved to the same row as the row to which the parent record of the table is moved. Data migration control method. 5. A data migration control program for causing a computer to execute each process in the data migration control method according to claim 4.