JPH04112246A - Derived data base processing system - Google Patents

Abstract

PURPOSE:To easily realize a form of operation by accessing plural data bases from one application program. CONSTITUTION:When the operation of simultaneous access from one application program to tables TA1, TB1, and TB2 is necessary, definitions of data bases 17A and 17B and a derived data base 18 are registered through a derived data base register part 11. When it is found in a data base operation part 12 that the operation object is the derived data base 18, it is converted to access information such as a storage structure related to a data actual body as constituting elements in accordance with register information of a dictionary 15. Thus, the application program sees the derived data base 18 as if, it were one data base having the actual body, and unific processing for plural data bases 17AS and 17B is possible.

Description

[Detailed Description of the Invention] [Summary] Regarding a derived database processing method in a database processing device in which a plurality of databases exist independently, application program operations that access multiple databases can be easily performed as if they were operated on a single database. The purpose is to provide the means to achieve this.

A dictionary that stores database definition information,
Database selection information 2 to be associated with the derived database A derived database registration means for registering selection information, etc. regarding the components of the selected database in the dictionary, and when the operation target is the derived database, the dictionary is referred to and the necessary database information is registered. The system is configured to include dictionary information scraping means for selecting constituent elements and replacing aliases, and database processing procedure creation means for creating a database processing procedure for a partial collection of storage elements of a plurality of databases. do.

(Industrial Field of Application) The present invention relates to a derived database processing method in a database processing apparatus in which a plurality of independent databases exist in which data collected for a certain purpose is simultaneously shared by a plurality of users.

As requirements for database management systems tend to become more sophisticated, there is a need for technology that realizes a highly scalable and flexible database operation format.

[Conventional technology]

For example, the standard for the SQL language, which is known as a computer language for databases (for example, the database language S
Q L JrSX3005-1990) defines a database as a collection of all data defined by multiple schemas in one environment. Therefore, one application program accesses one database.

However, the usage pattern (environment) of the actual database
However, it is difficult to use only the framework of the concept of data access such as the SQL standard. That is, due to practical requirements, the following processing may be required across multiple databases.

■ Between independently developed databases ■ Between departmental databases and central databases ■ Between independently operated databases with the same schema structure ■ Between personal databases and shared databases ■ Between test data and production databases In an operation mode that accesses multiple databases, as appropriate,
The reality is that solutions have been adopted depending on the situation.

For example, when accessing data on a plurality of independently developed departmental databases, the following nine methods are conventionally used.

The first method is to integrate the defined schemas into one database and reconstruct it. This allows application programs to share data.

The second method is to use the defined schema as an independent database for each department. Application programs are created for each department, and data is exchanged via general files to link application programs between multiple databases.

The above measures have placed a heavy burden on database construction or application program development.

That is, the purpose of creating a database is to share data, eliminate data redundancy, and eliminate semantic contradictions between data. Therefore, in principle, it is conceivable to combine the defined schemas into one database. However, the degree of operational independence for each organization, development period,
Currently, databases have to be developed in stages due to the amount of development, and it is often not practical to combine all data into one database.

[Problem to be solved by the invention]

In the conventional technology, only one database can be handled at the same time from one application program. The first problem was that multiple databases could not be handled at the same time.

■ Cross-referencing tables, data movement, and data updates between independently developed databases
They cannot be handled from the same application program at the same time.

■ Between the departmental database and the central database.

Coordination between departments is extremely difficult.

■ When independently operating databases have the same schema structure, a single application program cannot handle data contained in tables with the same definition that are distributed across multiple databases.

■ Personal and shared databases cannot be handled simultaneously from one application program, which has a negative impact on operability.

■ Between the test data and the actual operation database, the test data is used as a database and is mixed with the actual data cache, and it is not possible to perform a test using the actual data cache.

In addition, in order to be able to handle multiple databases in practice, it is necessary to prevent duplicate names from appearing in multiple databases. There is a problem in that the two individual databases cannot be developed independently.

SUMMARY OF THE INVENTION The present invention aims to solve the above-mentioned problems and provides a means for easily realizing application program operation that accesses a plurality of databases as if it were an operational form divided by one database.

[Means to solve the problem]

FIG. 1 is a diagram explaining the principle of the present invention.

In FIG. 1, 1 is a processing device consisting of a CPU and memory, etc., IO is a database definition unit that interprets and registers database definition statements, 11 is a derived database registration unit,
Reference numeral 12 denotes a database operation unit that processes access requests to the database, etc.; 13, a dictionary information operation unit; 14, a derived database interpretation unit; 15, a dictionary that stores definition information of the database and derived database; 16, a database storage device; 17A17B
1 represents a database, and 1st represents a derived database 19 representing a database processing procedure creation unit.

Databases 17A and 17B are independent databases. In the present invention, a partial collection of these database components can be defined as the derived database 18.

The derived database 18 can be interpreted as a "derived database" without breaking the definition range of names handled in the schema definition that defines the data contained in the database.
This is a new framework that defines a name, and in terms of semantic rules, it specifies processing for referential constraints in the schema definition related to the table selected in 5.

Derivation database registration unit 11 of database definition unit 10
inputs such a definition sentence in the derivation database 18 and performs a process of registering the definition information in the dictionary 15.

The dictionary 15 is a storage device that collectively manages information about schema definitions for reading this database and database storage devices on secondary storage. The dictionary 15 itself can also be constructed as one database.

Definition information registered in the Dikun Yonari 15 includes, for example, database selection information 3 that corresponds to the derived database 18, selection information regarding the components of the selected database, and alias designation information for changing names handled by the database to other names as necessary. is included.

When there is an access request to the databases 17A, -17B or the derived database 18, the database operation unit 12 generates an access schedule for processing the access request.

Here, when the operation target is the derivation database 18, the dictionary information operation unit 13
5, the derived data Hose solution station part 14↓Thus,
When a logical storage element is selected for the necessary data, if an alias is specified, the alias is replaced with the original name. In addition, the database processing procedure creation unit 19 performs an optimization process that specifically determines a database access route.
A database processing procedure for a partial collection of storage elements of a plurality of databases is created based on information regarding derived databases registered in the dictionary 15.

As a result, the derived database 18 is presented to the application program through the seven data processing procedures as if it were one real database.

[Effect]

The basic idea of the present invention is to introduce the idea of a database view to a database on multiple databases, similar to the relationship of a view to a table in a relational database.

For example, database 1τA or table TAI T
A2. Suppose that it is composed of TA3. Also, the database 17B has two tables TB1. It is assumed that it is composed of TB2TB3.

In these databases, one application program stores table TA1. If it becomes necessary to access TBI and TB2 at the same time, 5 Derived data base (C) 1
As database 17A, database 17B and their constituent tables TAI, TB1. T.B.
2 is registered in the derived database 18 through the derived database registration unit 11. In this definition.

It is possible to give a name different from the original name to the name of a table, etc. handled in the derived database 18, so that if the name of a table, etc. in the database 17A and the name of a table, etc. in the database 17B overlap. Even in such cases, collisions between one name can be avoided by giving one name an alias.

When the database operation unit 12 finds that the operation target is the derived database 18, the dictionary 1
According to the registered information in step 5, the conversion into access information such as the storage structure regarding the data entities that are the constituent elements associated with the derived database 18 is performed, so that from the perspective of the application program, the derived database 18 is a single entity. It looks like a database, and unified processing can be performed on the plurality of databases 17A and 17B.

〔Example〕

First, in order to clarify the concept of the derived database according to the present invention, we will introduce a solution based on the unified idea of 5 derived data hese for the operation of multiple databases that is practically unavoidable. Explain by comparing.

(, 1) Between independently developed databases There is an operational form in which databases are independent for each department. Independence means that database design, application program development, database management, and execution are performed by each department.

Even in this type of operation, there may be cases where it is desired to access data across data in other departments.

In the conventional method, in such cases, the defined schema is
By integrating the data into one database and allowing application programs to share data, or by using the defined schema as a database for each department, creating application programs for each department, and passing the data via a general file.
Methods such as data transfer and processing were adopted.

In contrast, the derived database can take a first-order approach.

1) Database design is done unrelatedly between independent departments, so even if the same schema name or title name exists, it is often extremely difficult to mutually modify the names. By specifying a unique name, it is possible to provide a means to avoid the above-mentioned collision of the same name through the definition of the derived database.

ll) To other departments:! There are many cases where you want to access the necessary tables. In contrast, with a derived database, necessary tables can be selected from multiple databases.

FIG. 2 shows an example of the definition of an independently developed database and a derived database.

Banking operations that have an accounting system and a receivables transaction system may operate on a two-page database, as shown in Figure 2. Even in this case, it is necessary for the receivables transaction application program to inquire about the balance of the account table in the accounting database and to update the account table using the contents of the transaction details table in the receivables transaction database.

In this embodiment, the account table of the accounting database,
A delivery derivation database is defined from the receivable transaction database. An application program can access multiple databases by passing through this derived database.

(2) Between departmental database and central data base In addition to the central database, there is an operational form in which each department has an independent data base. Central database refers to a centrally managed database.

The departmental database is 2.0 compared to the central database.
It is local data, managed by each department, and has a large number of databases.

Even in this mode of operation, there may be cases where it is desired to access departments and the central database at the same time. There are three types of operation modes for simultaneous access. One is to extract data from a central database to a departmental database, and perform W collection and processing.

The second method is to store data from one department database in a central database. The third case is when multiple databases between two departments are handled simultaneously.

In the derived database, simultaneous access between the central database and the departmental database can be handled in the same manner as in (1) above. Particularly when dealing with databases for two departments at the same time, it is difficult to predict and avoid name collisions due to the independence of each department.

Figure 3 shows an example of two departmental databases and a central database.

For example, as shown in FIG. 3, common data, which is a rough logical structure regarding wholesale trade, relationships between manufacturers and wholesalers and retailers can be associated with the central database d1 and the departmental database d2. Common data If the manufacturer, wholesaler, and retailer are each databases, the relationship between the central and departmental databases corresponds to the relationship between the common and the manufacturer, the common and the wholesaler, and the common and the retailer. The relationships between departments can be between manufacturers and wholesalers, and between wholesalers and retailers.

In this example, the common data database schema is information related to other databases.

Manufacturer information, manufacturer product attribute information, wholesale store ledger, retail store ledger, consumer information (Household Accounting Survey 2 Demographics.

There are various statistics).

In the table definition of the schema of the manufacturer's database, information related to other databases includes the manufacturer's internal results and budget planning table.

Information related to other databases in the table definition of the schema of the wholesale database includes sales results.

Information related to other databases in the table definition of the retail database schema includes POS data and H77 data.

For data sharing between departments, the manufacturer department processes information for each product, the wholesale department processes information for each product, manufacturer, wholesaler, and retailer, and the retail department processes retail information.

For example, in order for a wholesale department to perform an analysis by product that it does not have the ability to handle in-house, the product analysis derivation database should include manufacturer information and manufacturer product attribute information from the central database, performance information and budget information from the manufacturer database, and the wholesale database. Select the sales results from the list and define the derived database.

Between the central and departmental databases is the manufacturer department.

There is data sharing for the wholesale and retail departments to obtain consumer information, product information, and various analytical information.

For example, when storing information in a central database, 1
To refer to information in the departmental database.

Define a derived database. Conversely, in order to obtain analytical information from a single department database, analytical information in a central database may be referred to.

(3) Independently operated databases with the same schema structure There is an operational form in which database design, application program development, and database management are concentrated, and databases are independent for each of the five departments.

Even with this type of operation, when an application program accesses multiple databases, data from tables with the same name may be aggregated.

In the conventional method, application programs independently extracted data from each database and merged the data later.

Since the derived database can handle the above-mentioned (1), there is no need to process multiple databases individually.

FIG. 4 shows an example of independently operational databases having the same schema structure.

In corporate databases, data may be managed independently even if the schema definitions are the same. In the example shown in FIG. 4, the sales database is divided into regions and operated. The main reports consist of sales information and customer information.

Normally, a database is handled for each region, but when looking for company-wide sales, it is necessary to access multiple databases at the same time. By defining derived databases in necessary tables, columns, etc. of such a sales database group, this request can be easily met.

(4) Between a personal database and a shared database In the relationship between a personal database and a shared database that stores data necessary for one individual's work.

There are two modes of operation. One is to extract data from the 9 shared database to a personal database.

This is a form of editing and processing. The other method is to refer to a shared database in order to process a personal database. The latter has the same form as (2) above.

To address these problems, in the conventional method, it is necessary to first take out data from the shared database, edit and process it in the 5 personal databases, and then return it to the 5 shared databases.

The derived database simplifies the processing procedure by allowing access to multiple databases at the same time.

FIG. 5 shows an example of three personal databases and a shared database.

The personal database includes customer information for one individual, and the shared database includes general customer information, product information, and the like. You can refer to regional product sales information, or retrieve new product information into your personal database to create a quote.

When reflecting customer information in a personal database to general customer information in a shared database.

Data exchange between the personal database and the shared database is required. This kind of data linkage.

By using a derived database, N simple implementations can be achieved.

(5) Between test data and real-speed database In test systems, tests are conducted using the real-world database and the test database at the same time.For example, when developing a new application program while the online system is in operation. etc.

In the conventional method, two methods can be considered. One method is to create a separate test database of the same size. This method requires up to twice as much energy. The other method is to add a test table with a different name to the actual database, test the application program with the different name, and then write back to the original name after testing.

In the derived database, apart from the actual database,
Create a table to be updated by the program under test in the test database and update it without changing the application program.
It is possible to use both databases at the same time.

FIG. 6 shows an example of the test database.

As shown in Figure 6, manufacturer information and wholesaler ledger.

If you want to test new manufacturer information while operating on a practical data base consisting of retail store ledgers, etc., and only refer to the wholesale store ledger and retail store ledger, the test database will be used as the derived database 18. By defining the database to include the manufacturer information and the wholesale store ledger and retail store ledger of the actual data base, the manufacturer information can be easily tested using the derived database 18.

By using the derived database as described above, it becomes possible to flexibly respond to various types of operation.

Deriving databases is similar to developing views to tables. It can be said to be a database level view.

As shown in FIG. 7, in the conventional technology, the application program 3
Each of them could only access one database, such as 0A to database 17A and application program 30B to database 17B.

In the derived database, the application program 30C
As shown in FIG. 1, necessary portions of the plurality of databases 1A17B can be accessed simultaneously as the derived database 18.

At this time, if the same table master exists in a plurality of databases, changing the table name to alias T1 makes it possible to uniquely identify the table master.

FIG. 8 is an explanatory diagram of registration of a derivation database according to an embodiment of the present invention.

Database registration unit 4 in database definition unit 10
2 interprets the normal database definition statement 40 and registers the definition information in the dictionary 15. Further, the derivation database registration unit 11.

The derived database definition statement 41 is interpreted and its definition information is registered in the dictionary 15.

The BNF format of the database definition statement 40 and derived database definition statement 41 is as follows.

9 In this embodiment, in order to separate the logical structure and the storage structure and store one data unit (table or a part thereof) of the logical structure representation in association with a plurality of storage management units, A storage unit (CS) is a basic component of a database that has an independent structure on a physical medium.
A management unit called "Composite 5Lstructure" is used. C3 is a storage structure created from a combination of basic data organizations such as heap, B-tree, and hash, and can be stored in one table in one schema, multiple tables, or a specific column in a table depending on other definition information. It is now possible to correspond. Needless to say, even without using the concept of C3, 2
It is possible to implement the invention.

[Database definition statement format] <Database definition>::= CREATE DB
<Database salt> [FROM <cs specification>] Database salt>::= <Identifier><csspecification>
::= <C3 name>[(・<csname>)...]'[< in the C5 definition of each O8 specified in the syntax rule <cs name>
The database salt must match the database salt in the database definition.

[Construct a database from each C3 specified by <cs name> in semantic rule code <cs specification>. This database is named with <database salt>. If <cs specification> is omitted, C5 that specifies the <database salt> in the database specification in the O3 definition becomes a component.

[Description format of derived database definition statement] <Derived database definition>::= CREATE VDB <
Derivation 'i database salt FROM <Element DB specification>
[WHERE <Replacement list> Koku derived database salt>::= <Identifier><Element DB specification>::= <
Database salt>[<cs list>] [(,<database name>[<cs list>ko)61,]
<CS list>::=(<cs name> [(, <cs name>
1. ,, ]) Replacement list>::・<Replacement element> [(, <Replacement element>1.,,]) Replacement element>::= <Schema name 1> AS
<Schema name 2> (IN <Database salt><Qi711 name 1> AS <Table name 2> 0\ <
Data <, -path name>: Syntax rules, 1) The data header specified by <path name to 't-9> in the element Dll specification> is defined in <database definition> or derived database definition>. There must be.

2) Each C5 specified in <CS list> is the corresponding CS
It must be explicitly or implicitly specified in the database definition or derived database definition of the data specified in the database list immediately before the list.

3) The schema or table name specified by schema salt l or table name 1 in <replacement element> is 1.
It must be a comprehension of the data hesis specified by the database salt.

4) Schema salt 2 or table name 2 in <replacement element> must be unique within the derived database definition.

[Semantic rule 1 1) Construct a database by extracting C3 specified by <C5 name> in <CS list> from the data base specified by <database salt> in <element DB specification>.

This data base is named with the derived database salt. <Omit the CS list 'ek''J means that all C8s of the specified data base are specified.

2) On the data base specified by <replacement element> to the data specified by -y, name>, the schema specified by <schema salt> (this is the inclusion of our data base)
Replace the identifier with schema salt 2>.

Similarly, on the data hese specified by the replacement element> inside database salt>, set the identifier of the table specified by <table name 1> (which is a connotation of the data hese) to <table name 2>. replace.

3) Identifiers are not replaced for schemas or tables that are not specified in the <replacement list>.

By inputting the data hese definition statement 40 or derived data hesis definition statement 41 according to such specifications, the following cubic information is stored in the dictionary 15.

(a) Data hese identifier such as database identifier b) Storage structural unit (CS) that constitutes the data hese
) Information such as the O3 identifier that identifies the derived database identifier and the selected data hess identifier (d) Information on the correspondence between the derived data hesse and the selected O3 (e) Schema, such as when there is a duplication of names between the data heses Correspondence information between aliases of salt or table names and real names Below, specific examples of data hash definition statements and the like will be given regarding an example of generating a product analysis derived data hash from a departmental data hese and a central data hash.

In this example, there are three data cache definition statements: the central data database, the manufacturer department database, and the wholesale department data database. This definition statement is written as follows.

C) IEATE DB Central DB FIIIOM (Manufacturer information CS, Manufacturer product attribute information CS,...) CREATE DB Manufacturer DB FROM (Results C5,
Budget CS,...) CREATE DB Wholesale DB F
The definition of the ROM (sales record CS,...) derived data hess identifier external information is as follows.

CREATE C5 Manufacturer C5TYPE B□TREE
F, ROM <Schema table name> ON DBO
LALLOCATE DB DBOI <DB area name> T, O Manufacturer information CS <DB area name> To
Manufacturer product attribute information Similar to CS, DBO2 is Manufacturer D
DBO3 is allocated to wholesale DB. Since the description of this definition statement is the same, a specific example will be omitted.

For example, the derived data head definition statement is written as follows.

CREATE VD, B Product analysis VDB FRO
MDBOI (Manufacturer information CS, Manufacturer product attribute information C3) DBO2 (Results CS, Budget C3) DBO3 (Sales results C5) Next, an example of the structure of the data hess identifier will be explained with reference to FIG.

The inputs of the binding control unit 50 shown in FIG. 9 are a derived database identifier or a database identifier and a module of the SQL standard. The module contains the schema salt and table name.

This schema name and table name may be aliases for derived data specification.

The input to the binding control unit 50 is the following command, for example.

BIND -m <SQL module file name>
n Product analysis VDB The contents of this <SQL module file name>
The QL statement relationship is described as follows.

5ELECT * FROM Manufacturer information, manufacturer product attribute information results, budget, sales results...Here, manufacturer information, ..., sales results are table names on the SQL schema.

The binding control unit 50 includes the dictionary information operation unit 13
The derived database identifier or data hash identifier is notified to the derived data hash interpretation section 14 in the.

The derived data heath interpretation unit 14 determines the elements of the dictionary 15 to be handled in the five-line binding process. The dictionary elements are a sequence of C3 identifiers of a storage structure definition, a sequence of schema salts of a logical structure definition, a sequence of table names included in the schema, real names and aliases of schema salts, and real names and aliases of table names.

The processing details are as follows.

) Obtain all data hese identifiers and C3 identifiers from the derived data hese identifiers.

1i) Find related table names from the C5 identifier. Find the associated schema salt from the table name.

i[i) Find an alias from the table name and schema salt.

The optimization processing section 51 inputs the schema name and table name to the logical information operation section 53 of the dictionary information operation section 13 and acquires logical information and storage information for optimization processing.

When the logical information operation unit 53 inputs the schema salt and table name to the derived data hese interpretation unit 14, it outputs the C3 identifier and the real name if it is an alias. The logical information operation unit 53 extracts logical information from table information. If there is a referential constraint defined in the SQL schema between tables, the logical structure information of that table is also extracted. As a result, when a table related to a referential constraint is updated, the referential constraint is maintained.

Further, the logical information operation unit 53 extracts meta information up to the physical information below the C3 information by inputting the C3 identifier to the stored information operation unit 54. If there are other tables in a reference relationship, their storage information is extracted.

The storage information operation unit 54 includes a storage structure information operation unit 55, a CS
Space information operation section 56. CS space/DB area conversion unit 57. It consists of a DB area information management section 58.

The storage structure information manipulation unit 55 manages information on data organization primitives that are elements of C3. The elements of C3 include
This includes basic data organizations such as heaps, B-trees, and hashes.

The CS space information operation unit 56 manages space information allocated to cs. The cs space/DB area conversion unit 57 has a function of associating a CS space with a collection of extent information called a DB area. D
The B area information management unit 5.8 manages extent information belonging to the DB area. This extent information includes volume information on the physical medium and extent information within the volume.

The storage information manipulation unit 54 determines a program for manipulating the data organization related to the storage structure included in O3. Next, C
S space information operation section 56. CS space/DB area conversion unit 57. The DB area information management unit 58 extracts information and determines the extent information of the volume to be actually accessed.

Based on these results, the optimization processing unit 51 outputs the input database processing procedure 52 so that the processing cost for the data storage is the best based on the logical information and storage information.

It is also possible to carry out the third order.

, the division of roles within the dictionary information operation section 13 may be changed. For example, the optimization processing section 51 may call the derivation database interpretation section 14.
When calling the logical information operation unit 53 from the ``Scheme Name'' and ``Chiefle'' names, you may also input two derived data hese identifiers or data hese identifiers.

II The dictionary 15 may be a table of a relay database or a directory having a unique control table structure.

■The derived data hese information on dictionary 15 is
It only needs to exist on the data directory that can be accessed during binding. If the system is physically distributed, it is only necessary to make one redundant arrangement or update-related change using one distributed data storage technique.

i. Binding may function before or during the execution stage of accessing the data cache.

FIG. 10 shows an example of an operational configuration at the time of execution regarding an application program 60 that includes the data hash processing procedure 52 created in one or more ways.

For example, the objects of the application program 60 written in a host language such as C language or COBOL language are combined with the data hash processing procedure 52. Binding methods include advance binding using a linkage editor and dynamic binding at runtime.

In addition, the application program 60 and the data access section 6
There is a string control section 61 that controls the connection between the two.

The data access unit 62 performs data access processing procedure 5.
Input/output to the data storage device 16 is performed according to the data storage structure to be accessed, DB space information, etc. determined by 2. Note that the example shown in FIG. 10 is just an example, and the configuration can be similar to the conventional technology.

〔Effect of the invention〕

As explained above, according to the present invention, a plurality of independent data bases can be accessed from a single application program, thereby making it possible to flexibly deal with increasingly complex and sophisticated user operation patterns. Since multiple data fields can be handled simultaneously from one application program, the following effects can be obtained, for example.

■ Between independently developed data bases, cross-referencing of tables between data bases, data movement, and pig updates can be handled from the same application program.

■ Generally speaking, it is extremely difficult to coordinate between departments and central data, but with the one-name replacement function, you can simultaneously handle inter-departmental data from one application program. Data reference and data movement between one department and central data can be done at the same time. It becomes possible to handle it from one application program.

■ Between independently operated data heirs with the same schema structure, a single application program can handle data in tables between independently operated data heirs that are included in tables with the same definition distributed across multiple data heirs.

■ Between personal and shared data bases, one individual's pig base and common data base can be handled simultaneously from one application program.

■ Between the test data and the actual database: The actual database can be used for testing by making the test data into a test database, mixing it with the actual database, and accessing it for reference from one application program. You will be able to do this.

In addition, since it is possible to select any table from multiple data bases for the derived data base, by making only the selected table accessible for the derived data base,
Accidental data destruction can be prevented.

Furthermore, even if there are duplicate names among the five data fields,
By defining the derived data header, you can give it an alias, so you can use it without modifying the existing data header.
This has the effect that the range of data that an application program processes can be easily expanded.

[Brief explanation of drawings]

Figure 1 is an explanation of the principles of the present invention. Figure 2 is an example of the definition of a derived data base according to the present invention. Figure 3 is an example of a departmental data base and a central data base. Figure 4 is an example of an independently operated pig base with the same schema structure. Figure 5 is an example of a personal data base and a shared data base. Figure 6 is an example of a test data base. Figure 7 is a diagram explaining the role of the derived data base according to the present invention. Figure 8 is an example of a derived data base according to an embodiment of the present invention. Registration explanatory diagram. FIG. 9 shows an example 5 of the configuration of a data storage operation system according to an embodiment of the present invention. FIG. 10 shows an example of an operational configuration at the time of execution according to an embodiment of the present invention. In the figure, 1 is a processing device, 10 is a data header definition unit 211
12 is a derived data base registration unit, 12 is a derived data base operation unit,
13 is a dictionary information operation section, I4 is a derived data hese interpretation section, 15 is a dictionary, 16 is a data hese storage device, 17A and 17B are data hese, 18 is a derivation database, 19 is a data hese processing procedure creation section,
TA1-TA3. TBI to TB3 represent tables.

Claims (1)

[Scope of Claims] In a database processing device in which a plurality of independent databases exist that simultaneously share data collected for a certain purpose with a plurality of users, logical definition information of the independent databases and storage of the database in secondary storage are provided. A dictionary that manages information (
15), it is possible to define a new name definition frame without breaking the definition range of name names handled in the schema definition that defines the data included in the database, and to maintain consistency in the semantic relationships between tables included in the database. Derived database registration means (11) for registering in the dictionary selection information of a database to be made to correspond to the derived database (18) that maintains the characteristics, selection information regarding the constituent elements of the selected database, and alias designation information for changing the name name handled in the database to another name. ), and when the operation target is a derived database, refer to the dictionary and select the necessary database components, and if an alias is specified, change the alias to the original name. a dictionary information manipulation means (13) that replaces storage elements of a plurality of databases based on information about derived databases registered in the dictionary in a binding process that performs an optimization process that specifically determines a database access route; a database processing procedure creation means (19) for creating a database processing procedure for data collected in a database, and accesses a plurality of databases from one application program.