JPH1063554A - Database configuration information management system - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To perform a search on data stored in a database, conventionally, it is necessary to explicitly specify a condition for generating a predetermined access method for a certain processing request. could not. The present invention makes it possible to explicitly specify conditions for obtaining an optimum access method for a predetermined processing request. A database configuration information management system according to the present invention includes a first catalog information defining a configuration of data stored in a database storage unit and a first catalog information such that an access method is more optimal. With the second catalog information created by changing the copy unit, a file name storing catalog information to be referred to in response to a predetermined processing request is specified in the catalog reference information 23, and the compiler 4 For the processing request, an optimum access method is obtained by referring to the second catalog information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention makes it possible to register a plurality of pieces of catalog information defining the structure of data stored in a database in accordance with a processing request for accessing the database. The present invention relates to a database configuration information management system capable of obtaining a simple access procedure.

[0002]

2. Description of the Related Art Generally, when performing access processing to a database, definition information defining the structure of data stored in the database is managed in the form of a data file called a catalog. This catalog is often represented as a data file in a database. As an example of such a database configuration management method, "ISO / IEC 9075 Dataaba"
se LanguageSQL ”,“ 21 Infor ”
motion Schema and Definit
ion Schema ”states,“ The base table in the database is
It is managed in a table in the database. " FIG. 17 is a conceptual diagram of a conventional database configuration management method. Reference numeral 5 denotes a computer main body, reference numeral 20 denotes a database storage unit for storing data, and the database storage unit 20 stores a database A21 and a database B22. Reference numeral 1 denotes a first catalog information storage unit that manages definition information that defines the structure of data stored in the database storage unit 20, 7 and 8 denote processing requests to the database storage unit 20, 28, 29 indicates an access method, 6 indicates an input unit for giving a processing request to the computer main body, 3 indicates a processing request analysis unit for analyzing the processing request given from the input unit 6, and 4 indicates processing requests 7 and 8. 2 shows a compiler for obtaining access methods 28 and 29 for. In this conventional example, the processing request analysis unit 3 is an SQL processing system (SQL is a structured query).
It is assumed that the database storage unit 20 is accessed by SQL.

Next, the operation will be described. When the processing requests 7 and 8 for the database are input to the computer 5, the processing requests 7 and 8 are processed by the processing request analysis unit 3 provided in the computer 5.
In the SQL processing system, based on the definition information stored in the first catalog information storage unit 1, processing is performed so that the access methods 28 and 29 are more optimal for the processing target database file. In the normal processing request analysis unit 3, the compiler 4 accesses certain processing requests 7 and 8 according to the definition information stored in the first catalog information storage unit 1 so that the compiler 4 has a more optimal access method. Generate schemes 28 and 29. However, even if it is possible to obtain the optimal access method for a certain processing request,
An optimal access method is not always obtained for another processing request. Further, the user cannot explicitly specify a specific access method.

For example, it is assumed that a data file called _COLUMNS which manages column definition information 11 as shown in FIG. 18 is stored in the first catalog information storage unit 1. The column definition information 11 is a schema name 11a, a column name 11c, and a table name 1 of a table in which the column exists.
1b, data type 11d, size 11e, and index information 11f as constituent elements. The column name 11c has an index 12. In FIG. 18, a thick frame is attached to CL_NAME to indicate the index 12. For example, a processing request for searching for a table name having a predetermined column is input as a processing request A, and a processing request for outputting a column of a predetermined table to a list is input as a processing request B from the input unit 6. The optimum access method is obtained in the section 3. S of processing request A and processing request B in this case
The QL statement is shown below. First, SQ of processing request A
Indicates an L sentence.

[0005] SELECT TB_NAME FROM_COLUMNS WHERE CL_NAME = ‘CL1’

The processing request A has a column name of “CL1”.
Is a processing request for searching for a table. Next, the processing request B is shown below by an SQL statement.

SELECT CL_NAME FROM_COLUMNS WHERE TB_NAME = ‘T1’

The processing request B described above has a table name of '
This is a processing request for searching for the column T1 '.

An index 12 is attached to the column name 11c to be searched by the processing request A. For this reason,
Since the column name 'CL1' can be searched based on the index 12, an optimal access method 28 can be obtained. However, column definition information 1 called _COLUMNS
When trying to obtain the access method 29 for the processing request B using No. 1, the compiler 4 generates an access method for sequentially accessing the table names, since TB_NAME is not indexed.

[0010] Japanese Patent Application Laid-Open No. 5-108724 is given as an example of a system for data access. In JP-A-5-108724, an access program for a database, that is, a processing request described in SQL or the like is registered, and the registered access program is used as a model. In this case, in the registered access program, an access method based on predetermined definition information is applied to all processing requests, and therefore, only an access method that is not optimal may be obtained, and a more optimal access method may be obtained. No method is described.

[0011]

In the conventional processing method,
As described above, a compiler generates an access method according to catalog information that is definition information for managing a database for a certain processing request. For this reason, a user cannot explicitly specify a specific access method, and an optimum access method cannot be obtained depending on a processing request, so that processing efficiency may be deteriorated. was there.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems. One catalog information can be defined by defining another catalog information and a specific access method can be designated, and the processing can be speeded up. It is intended to realize.

[0013]

A database configuration information management system according to the present invention stores a database storage unit for storing data, and first catalog information describing the configuration of data stored in the database storage unit. A first catalog information storage unit, an input unit for inputting a request for processing of data stored in the database storage unit, and a processing request input from the input unit to the first catalog information storage unit. In a database configuration information management system including a processing request analysis unit that obtains a predetermined access procedure for accessing the database by analyzing according to the stored first catalog information, the database configuration information management system includes a predetermined processing request A second catalog information storage unit for storing second catalog information created for The input unit inputs catalog reference information that specifies that the processing request analysis unit refers to the second catalog information stored in the second catalog information storage unit. The processing request input from the input unit is analyzed in accordance with the second catalog information stored in the second catalog information storage unit specified in the catalog reference information to obtain a predetermined access procedure for accessing the database. It is characterized by the following.

Further, in the database configuration information management system according to the present invention, the second catalog information storage unit changes the first catalog information stored in the first catalog information storage unit to predetermined catalog information. And store it as second catalog information.

Further, in the database configuration information management system according to the present invention, the second catalog information storage unit changes a part of the first catalog information stored in the first catalog information storage unit. Then, the changed part of the information is stored as second catalog information.

Further, the database configuration information management system according to the present invention is characterized in that the second catalog information storage section stores catalog information in a predetermined file format.

Further, the database configuration information management system according to the present invention is characterized in that the input unit inputs the processing request and the catalog reference information as one piece of information.

In the database configuration information management system according to the present invention, the first catalog information is composed of one or more fields, and the first catalog information storage unit stores the first catalog information as the first catalog information. First index information indicating whether the field is an indexed field or a non-indexed field is stored for each of the fields, and the second catalog information includes one or more fields.
The second catalog information storage unit stores, as the second catalog information, second index information indicating whether the field is an indexed field or an unindexed field for each of the fields,
When the first catalog information storage unit stores the first index information indicating that the predetermined field is an indexed field as the first catalog information, the second catalog information storage unit stores For the predetermined field, the second index information indicating that the field is a field without an index is stored.

Also, in the database configuration information management system according to the present invention, the first catalog information storage unit stores the first catalog information as the first catalog information for each of the fields.
And the first field name indicating the field is stored, and the second catalog information storage unit stores the second index information for each field as the second catalog information. And the second field name indicating the field is stored, and the first catalog information storage unit stores the first field name and the first index information for each of the two or more consecutive fields. The second catalog information storage unit stores the second field name and the second index information with the two or more consecutive fields as one field.

Further, in the database configuration information management system according to the present invention, the first catalog information storage section stores a first data length indicating a length of data stored in the field as the first catalog information. And stores a first index length indicating that a part of the data stored in the field is indexed, and stores the second catalog information in the field as the second catalog information. The first catalog information storage unit stores a second data length indicating the length of data and a second index length indicating that a part of the data stored in the field is indexed. When the first index length in which a part of the data is indexed for the field having the first data length is stored, the second catalog information storage unit includes: And indexed the first data length with respect to the serial above field is the first data length and to store the second index length.

Further, in the database configuration information management system according to the present invention, the first catalog information storage section stores a first field name indicating the field as the first catalog information, An information storage unit stores a second field name indicating the field as the second catalog information, and the first catalog information storage unit stores the first field name for two or more consecutive fields. When the information is stored, the second catalog information storage unit stores the second field name with the two or more consecutive fields as one field.

Further, in the database configuration information management system according to the present invention, when the first catalog information storage section stores the first field name for at least one of the fields, the second catalog information storage section Is characterized in that the one field is divided into two or more fields, and two or more second field names are stored.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 FIG. Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a configuration of a database configuration information management system according to the first embodiment. In FIG. 1, the database configuration information management system changes a first catalog information stored in the first catalog information storage unit 1 to a second catalog created by changing a part of the first catalog information. The figure shows that a second catalog information storage unit 2 for storing information is provided. Also,
The computer 5 includes a first catalog information storage unit 1 for storing first catalog information for managing data stored in the database storage unit 20, and a second catalog information storage unit for storing second catalog information. 2 and a processing request analysis unit 3 having a compiler 4. Reference numeral 6 denotes a terminal or the like for inputting processing requests 7 and 8, and is connected to the computer 5. Reference numeral 23 denotes catalog reference information for specifying a file storing catalog information to be referenced by the compiler 4 in response to a predetermined processing request, and is input from the input unit 6. In the first embodiment, data stored in the database storage unit 20 is accessed using an SQL statement, as in the conventional example. For this reason, the processing request analysis unit 3 becomes an SQL processing system. FIG.
FIG. 4 is a diagram illustrating an example of first and second catalog information stored in first and second catalog information storage units. FIG. 2A shows the first catalog information. The first catalog information is a data file called _COLUMNS that manages the column definition information 11. The column definition information 11 includes a schema name 11a, a column name 11c, a table name 11b in which a column exists, a data type 11d, a size 11e, and index information 11f. The column name 11c has an index 12. FIG. 2B shows the second catalog information. The second catalog information is column definition information 13
Is a data file called _COLUMNS. The column definition information 13 includes a schema name 13a, a column name 13c, a table name 13b of a table in which the column exists, a data type 13d, a size 13e, and index information 13f. Also, the table name 13b
Has an index 14.

Next, the processing procedure in the database configuration information management system will be described with reference to the flowcharts of FIGS. FIG. 3 is a flowchart showing a procedure performed by the user.
FIG. 4 is a flowchart showing the operation of the compiler. As shown in FIG. 3, the user can first obtain the optimum access method by referring to the processing requests 7 and 8 and the first catalog information stored in the first catalog information storage unit 1. It is determined whether it is (S15). If it is not possible to obtain the optimum access method in S15, the second catalog information 2 to be replaced with the first catalog information is created (S15).
16). The second catalog information is created by changing a part of the first catalog information. However, the second catalog information includes information that has not been changed in the first catalog information, and is referred to as second catalog information. In the first embodiment, it is assumed that the processing request 7 refers to the first catalog information and the processing request 8 refers to the second catalog information. Therefore, the processing request (SQL sentence) 8 is rewritten as necessary so as to be suitable for the catalog information 2 (S17). In S16, the second catalog information is created by changing a part of the first catalog information. For example, FIG.
In the first catalog information shown in (A), column name 1
Index 12 was attached to 1c. However, the user cancels the index 12 attached to the column name 11c as shown in FIG. 2B and obtains an index 14 for the table name 13b to obtain the second catalog information as the second catalog information in order to obtain the optimum access method. Is stored in the catalog information storage unit 2. For this reason, for example, when the processing request 8 is a processing request that outputs column names in ascending order of table names, the processing request 8 includes the acquired column names in ascending order of table names corresponding to the column names. The processing for rearranging is included. However, as shown in FIG.
If the table name and the column name are acquired based on the index 14 by adding, the acquired column names are arranged in ascending order of the table names. For this reason, the process of rearranging the column names for each table name defined in the processing request 8 becomes unnecessary, and the processing request 8 needs to be rewritten. Next, at the time of compiling, catalog reference information 23 (hereinafter referred to as a make file) is created which designates the processing request 8 to refer to the catalog information 2 (S1).
8). FIG. 5 shows an example of a make file. In the make file 23 shown in FIG. 5, for the processing request stored in “SQL1.SQL”, the access method is obtained by referring to the file “CAT02” in which the second catalog information is stored. Instructed. FIG. 6 shows an example of another make file. The make file 23 shown in FIG. 6 shows an example in which the contents of the processing request 8 are directly described in the make file 23. As shown in FIG. 6, the processing request may be described together with the file name storing the catalog information to be referred to in the make file. Thereafter, the user inputs processing requests 7 and 8 from the input unit 6.
And the catalog reference information 23 to the computer 5 (S1
9). In the computer 5, the compiler 4 generates the access methods 9 and 10 by referring to the first catalog information or the second catalog information according to the catalog reference information 23 (S2).
0). Finally, the computer 5 outputs the more optimized access methods 9 and 10 (S21).

The operation of the compiler 4 in S20 will be described below with reference to the flowchart of FIG. First, the compiler 4 refers to the make file to obtain a file name in which catalog information to be used is stored (S3).
1). Next, it is confirmed whether a text file having the acquired file name exists (S32). The text file refers to MS-DOS (Microsoft-Disk O
The operating system, MS-DOS, is a file in a file format accessible by Microsoft Corporation (registered trademark). Then, if a text file exists, the text file is read (S
33). The text files appearing in S32 and S33 will be described in detail in the following embodiments. S32 above
In step (3), if there is no text file, it is checked whether a subset of catalog information exists (S3).
4) If a subset exists, the subset is read (S35). The subset is a file storing the second catalog information. If there is no subset, the first catalog information is read from the first catalog information storage unit 1 (S36). After that, S33 or S
The physical configuration information of the database to be searched is obtained from the file read in 36 (S37), and the process proceeds to a process of actually creating an access method. Next, Compiler 4
Based on the configuration information obtained in S37 and the processing request transmitted in S19, whether or not an item (column) serving as a sort key for executing a processing request in a table indicating a database to be processed has an index. Is determined (S38). If there is an index, the compiler 4 creates an access method using the index (S39). It is checked whether the column definition of the item (column) used as the sort key in the table indicating the database to be processed matches the index definition (S41), and if they match, an access method is created using the index. (S42). S38 above
If no index is assigned in step (S40), an access method that does not use an index is generated (S40). In addition, the above S41
If the column definition does not match the index definition, an access method that does not use an index is generated (S43). After the processing of S31 to S43 as described above, the column offset and the size of the item (column) used to execute the processing request are obtained (S44). The access method created in S40 or S42 or S43 includes an item (column)
Since no column offset or size has been set for, the optimal access method is obtained by setting the column offset size acquired in S44 (S45).

In FIG. 4, S38 and S4 are used as criteria for determining whether or not to use an index when creating an access method.
1 was taken as an example. However, another determination, such as whether the column definition of the table (column) indicating the database to be processed matches the column definition of the item (column) described as the selection condition in the processing request Standards are also conceivable.

Conventionally, when the compiler 4 generates an access method, if the first catalog information stored in the first catalog information storage unit 1 is used, the compiler 4 can optimize the processing request A for the processing request A. Although the access method 9 is generated, the optimum access method 10 cannot be generated for the processing request B. However, another second catalog information is prepared for the processing request B and stored in the second catalog information storage unit 2, and the processing request B is stored in the second catalog information storage unit 2. By inputting the catalog reference information 23 specified to refer to the existing second catalog information, the compiler 4 refers to the second catalog information and generates an optimal access method 10. For example, assume that there is a database that stores 100,000 data. When the search process is performed on the database, the unnecessary data becomes 1
It is assumed that there are no known cases. In this case, in the method described in the conventional example, for example, if an item (column) to be searched has an index as the first catalog information, an access method for accessing the database with reference to the index is created. For this reason, it takes time to access the database if the destinations of the addresses are different according to the index. However, by using the database configuration information management system of the present invention, a user can remove an index attached to an item (column) to be searched according to a processing request. Then, the changed first catalog information is registered as second catalog information, and if a certain processing request is specified in the catalog reference information so as to refer to the second catalog information, the compiler performs predetermined processing. For the request, the access method is obtained by referring to the second catalog information. For this reason, when searching for predetermined data from the above 100,000 database, if an item (column) to be searched is not indexed, it is determined whether or not the data matches the search condition in order from the first data. Obtain such an access method. Therefore, it is possible to obtain an access method with higher processing efficiency than an access method that refers to the index.

For example, the processing request B for outputting the column name list shown in the conventional example is TB_NA of _COLUMNS.
A second index obtained by adding an index 14 to the ME as shown in FIG.
Prepare catalog information for processing request B and compile 4
Specifies the catalog reference information 23 so that the processing request B refers to the second catalog information.
An optimal access method can be obtained. Then, high-speed database processing can be performed.

Embodiment 2 In the first embodiment, information obtained by partially changing the first catalog information and information that has not been changed are stored in the second catalog information storage unit 2 as second catalog information created for the processing request B. I remembered.
However, in the second embodiment, an example will be described in which only information obtained by partially changing the first catalog information is used as the second catalog information. FIG. 7 is a system configuration diagram of the database configuration information management system according to the second embodiment. The difference from the first embodiment is that only the changed information in the first catalog information is used as the second catalog information 36 in the second catalog information.
Is stored in the catalog information storage unit 2.

That is, in the first embodiment, the first catalog information storage in which the first catalog information is stored in order to obtain the access method 9 that optimizes the processing request A and the access method 10 that optimizes the processing request B There was provided a second catalog information 2 storing a first catalog information and a second catalog information. The second catalog information stores both changed information and unchanged information among the first catalog information. Therefore, the first catalog information storage unit 1 and the second
The definition information stored in the catalog information storage unit 2 is enormous information.

The system configuration of the database configuration information management system according to the second embodiment is the same as that of the first embodiment except that the change-target catalog information 1a is changed to the second catalog information. Is omitted. Also, in FIG. 3, the processing flow in the database configuration information management system is such that when the second catalog information is created (S16), not all the information but the change target catalog information 1a is changed and the second catalog information is changed. 3
Create 6. Further, regarding the processing flow of the compiler 4 of FIG. 4, the processing of checking whether the catalog subset corresponds to the second catalog information 36 and the catalog subset exists in S34 of FIG. 4, and the processing of reading the catalog subset in S35. Corresponds to the second embodiment described below.

For example, in the first embodiment, FIG.
_COLUM which is the column definition information 11 shown in FIG.
NS had an index 12 on the column name. Figure 2
As shown in (B), the column definition information 13 in which the index 14 is added to the table name and the index 12 of the column name is canceled is stored in the second
Is stored in the second catalog information storage unit 2 as the catalog information. The second catalog information storage unit 2 stores, in addition to the column definition information 13, other definition information which is not changed and is stored in the first catalog definition information storage unit 1. However, in the second embodiment, it is possible to store only the catalog information in which the index 14 is added to the table name as the second catalog information 36 in the second catalog information storage unit 2. When a processing request such as a processing request A or a processing request B is input from the input unit 6, the first processing request A is stored in the catalog reference information 23.
Compile 4 can obtain an optimal access method by designating to refer to the second catalog information 36 for the processing request B. Further, for a processing request that refers to the first catalog information, the first catalog information may be referred to as a default value without specifying the catalog information to be referred to in the catalog reference information 23.

As described above, since it is not necessary to duplicate information that matches between the first catalog information and the second catalog information, it is possible to reduce the data capacity.

Embodiment 3 In the first and second embodiments, the second catalog information is also provided as a data file group of the database. However, for example, it is also possible to provide as a text file. FIG. 8 is a diagram showing a system configuration of a database configuration information management system according to the third embodiment. 8 and FIG. 7 of the second embodiment.
Is that the file format for storing the second catalog information is not a database format file,
MS-DOS (registered trademark of Microsoft Corporation, Mi
cross-Disk Operating Sys
tem) is a text file 37 that is more accessible.

The third embodiment has the same system configuration as the first and second embodiments except that the second catalog information 37 is a text file. Further, in the processing flow, in the processing procedure of the user (FIG. 3), a text file is generated in the processing of creating the second catalog information (S16). In the processing flow showing the operation of the compiler 4, the processing of S32 of FIG. 4 for confirming whether there is a text file and the processing of S33 of reading a text file correspond to the third embodiment. If the file name of the catalog information to be referred to in the make file as the catalog reference information 23 is specified in the text file name, the compiler 4 refers to the second catalog information 37 which is the specified text file name and optimizes it. Access method.

For example, _COLUMNS which is the column definition information 13 shown in FIG. 2B is described as follows and stored in a text file. PATH = / CATALOG / _COLUMNS COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 COL = ■ CL_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 64:::::: As described in the conventional example, catalog information is often represented as a data file in a database. Therefore, when adding or changing catalog information, a tool having a function of accessing a data file in a database and editing data is required.
Also, the user had to know how to use the tool. However, as described above, the second catalog information created by changing a part of the first catalog information is stored in a text file, so that new catalog information is provided or the management information is changed. In this case, addition and change can be easily performed by a tool having a function of accessing and editing a text file. For example, if a user does not know how to use a tool that has the function of accessing a data file in a database and editing data, but knows a tool that has the function of accessing and editing a text file Therefore, the user can create catalog information according to the processing request.

In this way, a plurality of pieces of catalog information can be prepared more easily, so that it is possible to easily respond to various processing requests. Also, an example has been described in which the format of the file storing the second catalog information is a text file accessible from MS-DOS. However, the file format is not limited to a text file, and may be a file format accessible from a predetermined operating system.

Embodiment 4 FIG. Embodiments 1, 2, and 3
Then, the second catalog information that changes a part of the first catalog information and stores both the changed information and the unchanged information, or the second catalog information that stores only the changed information By having multiple according to the processing request,
Had the best access method. In the fourth embodiment, a description will be given of obtaining a more optimal access method by changing the way of viewing the table for managing the first catalog information.

As an example of changing the view of the table, there is an example in which definition information to which an index is added is defined as no index. FIG. 9 shows an example of table definition information according to the fourth embodiment. The table definition information 38 in FIG. 9 is called _TABLES, and manages information such as table names, owned schema names, and the number of columns. An index 39 is added to SC_NAME, which is a schema name. If this is converted into a text format, it can be described as follows.

PATH = / CATALOG / _TABLES COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 Index = YES COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 :::::: The table definition information 38 described above is stored in the first catalog information storage unit 1 as first catalog information. In the third embodiment, an example in which the second catalog information is stored in the text file has been described. In the fourth embodiment, the first catalog information can be further stored in a text file.

For example, it is assumed that a processing request for outputting the following list of schema names is input from the input unit 6. SELECT SC_NAME FROM_TABLES When using the table definition information 38 in FIG. 9 in response to the above processing request, the compiler 4 generates an access method using the index 39. Therefore, if a processing request is executed using the above access method, there is a possibility that a large number of inputs and outputs to the storage unit storing the table definition information 38 may occur, causing a problem that the processing speed is reduced.

Therefore, as shown in FIG.
The second catalog information without the index 39 is created for the ME. The user changes the first catalog information stored in the text file to the second catalog information described below. Then, it is stored in a text file.

PATH = / CATALOG / _TABLES COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 Index = NO COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 ::::::

The second catalog information is SC_NA
The index for ME is changed from "YES" to "NO". The second catalog information is created, and the compiler 4 designates the catalog reference information 23 so as to refer to the second catalog information. Then, the compiler 4 sets the SC_NAME in the table definition information 38a.
Is regarded as having no index 39, and an access method for sequentially accessing the database in response to the processing request can be obtained. Therefore, there is no access to refer to the index, and the processing speed performance is improved. FIG. 11 shows an example of a make file according to the fourth embodiment. The catalog reference information 23, which is a makefile shown in FIG. 11, is used to generate an access method for a processing request stored in a file “SQL1.SQL”, and to generate second catalog information named “CAT02.TXT”. Indicates that the user refers to the stored text file. FIG. 12 shows an example of another make file according to the fourth embodiment. The catalog reference information 23 which is the make file shown in FIG. 12 shows an example in which the processing request 8 is described in the make file as it is. As shown in FIG. 12, it is also possible to directly describe the processing request in the make file.

Embodiment 5 FIG. In the fourth embodiment, the definition information to which an index is added is regarded as “no index”,
Although a more optimal access method has been obtained, in the fifth embodiment, as shown in FIG. 13, a method of using an index added over a plurality of columns is also conceivable.

The column definition information 11 _COLUMNS shown in FIG. 13 includes SC_NAME, TB_NAME,
An index 41 is provided over a plurality of columns of CL_NAME. Such column definition information 11 is described in the following text format, and is stored in the text file as the first catalog information.

PATH = / CATALOG / _COLUMNS COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 Index = YES IdexLen = 96 COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 Index = YES COL = ■ CL_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 64 Index = YES COL = ■ IDX_INFO ■ Type = ■ CHAR ■ Size = 32 Offset = 96:::::

For example, it is assumed that a processing request described below is input from the input unit 6 to the computer 5. SELECT IDX_INFO FROM_COLUMNS WHERE SC_NAME = 'SC1' AND TB_NAME = 'T1' AND CL_NAME = 'CL1' The above processing request is a processing request for searching index information from a schema name, table name, and column name having a specific value. . When the above-described first catalog information is used for this processing request, each item SC_NAME,
Even if CL_NAME and CL_NAME are each indexed, the compiler 4 employs the sequential access method because the above conditions are connected by AND conditions.

Therefore, as shown in FIG.
C_NAME, TB_NAME and CL_NAME are regarded as one item (column) 40, and SC_T
Name it B_CL. To realize this, the above-described first catalog information is changed to the following column definition information as the second catalog information. The second shown below
In the catalog information of, SC_NAME, TB_NAME and CL_NAME in the above-mentioned first catalog information are defined as one item SC_TB_CL, and further defined as indexed. Then, the defined second catalog information is stored in a text file.

PATH = / CATALOG / _COLUMNS COL = ■ SC_TB_CL ■ Type = ■ CHAR ■ Size = 96 Offset = 0 Index = YES IdxLen = 96 COL = ■ IDX_INFO ■ Type = ■ CHAR ■ Size = 32 Offset = 96:::: :::

The processing request is also changed as shown below according to the second catalog information. SELECT IDX_INFO FROM_COLUMNS WHERE SC_TB_CL = CONCAT ('SC1', 'T1', 'CL1')

By changing the processing request and referring to the second catalog information as described above, the compiler 4
An access method using an index of _TB_CL can be adopted.

Embodiment 6 FIG. In the fifth embodiment, an example has been described in which catalog information is changed such that an index added over a plurality of columns is indexed to one item (column). In the sixth embodiment, as shown in FIG. 14, an example will be described in which, when an index is added to YYMM which is a part of DATE, second catalog information in which the item DATE is divided into a plurality of items is created. .

In the first catalog information defining the structure of the data of the database to be processed, _TABL
The table definition information 38 named ES indicates the last change date (YYMMDD) in the item (column) DATE42.
Index 43 in YYMM of DATE 42
When the first catalog information is provided in a text format, the first catalog information is as follows. PATH = / CATALOG / _TABLES COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ DATE ■ Type = ■ CHAR ■ Size = 6 Offset = 32 Index = YES IdxLen = 4::::::

For example, as shown in SELECT TB_NAME FROM_TABLES WHERE 951001 <= DATE <951101, the table definition information 38 indicates the last change date (YY
When MMDD is managed by the item (column) name of DATE42, a processing request for searching the table last changed in October 1995 is issued. When the first catalog information is used for the processing request, DATE4
It is impossible to use the index 43 for No. 2 (since the index is attached to YYMM constituting DATE 42). Therefore, the compiler 4 obtains a sequential access method for performing the comparison process twice. In the first comparison process of the above two comparison processes, 951001 ≦ D
Perform ATE. DATE ≦ 9 in the second comparison process
The comparison processing of 51101 is performed. In addition, since the index 43 cannot be used, all data stored in the database is accessed.

Therefore, DATE 42 is regarded as being composed of two items (columns) of YYMM and DD. To realize this, the second catalog information is created in a text format as shown below. PATH = / CATALOG / _TABLES COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ YYMM Type = ■ CHAR ■ Size = 4 Offset = 32 Index = YES I dxLen = 4 COL = ■ DD ■ Type = ■ CHAR ■ Size = 2 Offset = 36::::::

The processing request is also changed as follows. SELECT TB_MAME FROM_TABLES WHERE YYMM = 9510 When the access method is generated with reference to the second catalog information for the processing request changed as described above, the compiler 4
Generates an access method that performs the comparison process using the index 43 once.

Embodiment 7 FIG. In the sixth embodiment, an example is described in which, when an index is added to a part of a certain item (column), the second catalog information is created by dividing the item (column) into items (columns). explained. In the seventh embodiment, as shown in FIG.
If a selection condition such as E and CL_NAME for determining a match for each of a plurality of continuous items is specified in the processing request, and the selection conditions are combined by a logical product, a plurality of continuous An example of creating the second catalog information in which the above items are combined into one item will be described below.

In the first catalog information that defines the data structure of the database to be processed, _COLU
TB_NAM of column definition information 11 named MNS
E and CL_NAME are separate columns
Is defined in the catalog information. An example in which the first catalog information is described in a text format is shown below. PATH = / CATALOG / _COLUMNS COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 COL = ■ CL_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 64 ::::::

For example, SC_NAME FROM CL_NAME = “CL1” and SC_NAME to SC_NAME, such as SELECT SC_NAME FROM_COLUMNS WHERE TB_NAME = “T1”
The user attempts to execute a processing request for searching for NAME. In response to the processing request, the compiler 4
When the access method is generated using the catalog information of
The second time, check if the value of TB_NAME matches the condition,
At the second time, it is checked whether the value of CL_NAME matches the condition. Therefore, two confirmation processes are required.

Therefore, TB_NAME and CL_NAME
Is created as one item (column) 44 called TB_CL. An example in which the second catalog information is described in a text format will be described below. PATH = / CATALOG / _COLUMNS COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ TB_CL ■ Type = ■ CHAR ■ Size = 64 Offset = 32:::::

The processing request is changed as follows in accordance with the above-mentioned second catalog information. SELECT SC_NAME FROM_COLUMNS WHERE TB_CL = CONCAT ('T1', 'CL1') The compiler 4 executes the confirmation processing only once by referring to the second catalog information for the processing request changed as described above. An access method can be generated.

As described above, an optimal access method can be obtained by forcibly adopting a processing method in which a plurality of selection conditions can be processed at a high speed as one selection condition.

Embodiment 8 FIG. In the seventh embodiment, when the selection condition for determining the match for each of a plurality of continuous items is specified in the processing request and the selection conditions are combined by a logical product, the continuous The method of combining a plurality of items as one item has been described. In the eighth embodiment, as shown in FIG. 16, when a selection condition for judging a partial match of the value of the above item with respect to the preceding or following item is specified in the processing request, An example will be described in which an item is defined such that an item is defined so that the value before or after the value of the item is a value for another item.

In the first catalog information that defines the data structure of the database to be processed, information such as table names, schema names, and the number of columns is managed in FIG.
Table definition information 38 named _TABLES is shown below in text format. PATH = / CATALOG / _TABLES COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 COL = ■ CL_CNT ■ Type = ■ INT ■ Size = 4 Offset = 64::::::

For example, the user tries to execute a processing request to display all schema names having a table name starting with “TB”, such as a processing request of “SELECT SC_NAME FROM_TABLES WHERE TB_NAME LIKE“ TB% ”. When the first catalog information is used for the processing request, a meta-character comparison is performed to confirm whether the first two characters of the value for TB_NAME are 'TB'.

Therefore, as shown in FIG. 16, in the table definition information 38, TB_NAME is set to A145 and A2.
The second catalog information is created so as to be regarded as an item consisting of 46 items (columns). Hereinafter, the second catalog information created using the first catalog information is shown in a text format.

PATH = / CATALOG / _TABLES COL = ■ SC_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 0 COL = ■ TB_NAME ■ Type = ■ CHAR ■ Size = 32 Offset = 32 COL = ■ A1 ■ Type = ■ CHAR ■ Size = 2 Offset = 32 COL = ■ A2 ■ Type = ■ CHAR ■ Size = 30 Offset = 34 COL = ■ CL_CNT ■ Type = ■ INT ■ Size = 4 Offset = 64 COL = ■ DATE ■ Type = ■ CHAR ■ Size = 6 Offset = 68::::::

Further, the above processing request is also changed as follows. SELECT SC_NAME FROM_TABLES WHERE A1 = 'TB' The processing request is changed as described above, and the changed processing request is input from the input unit 6 to the computer 5. Thereby, the compiler 4 generates an access method for performing a simple confirmation process.

As described above, instead of determining that the value for a certain item (column) partially matches, it is determined that the entire value for a certain item (column) matches, and processing can be performed at high speed. Force the system to be adopted.

[0071]

Since the present invention is configured as described above, it has the following effects.

In the database configuration information management system according to the present invention, in a database processing system, first catalog information for managing definition information defining a configuration of data stored in a database is changed in accordance with a predetermined processing request. To create second catalog information. Then, for a predetermined processing request, it is possible to explicitly force the access method to be obtained by referring to the second catalog information. Therefore, there is an effect that an optimal access method can be selected.

Further, since the first catalog information manages the definition information of all data stored in the database, the storage capacity becomes enormous. In the above invention, the catalog information that has not been changed among the first catalog information is also stored in the second catalog information storage unit as the second catalog information. However, the second catalog information can be only required partial information of the first catalog information, that is, only the catalog information to be changed.
For this reason, there is an effect that the capacity for storing the second catalog information can be made smaller than in the above invention.

The first catalog information itself is usually managed as partial information in the database. However, the second catalog information is not managed as partial information in the database, but can be provided in a file format that can be accessed by a predetermined operating system, for example, MS-DOS. For this reason, when creating a second catalog information by changing a part of the first catalog information, a user must know how to use a tool for accessing and editing information in the database. If the file format is accessible by MS-DOS, the data can be changed by a tool for accessing and editing the file in the MS-DOS file format, so that the catalog information can be easily changed. There is.

The input unit allows a processing request and catalog reference information to be input as one piece of information.
For this reason, when the user checks which processing request refers to which catalog information, since the processing request and the catalog reference information are in one file, it is easy to manage the processing request and the catalog reference information. There is an effect.

Further, a field to which an index is added in the first catalog information can be defined as an indexless field in the second catalog information. For this reason, a processing method that does not use an index can be forcibly adopted for a predetermined processing request, so that an optimum access method can be selected.

When the added index is defined over a plurality of fields in the first catalog information, the plurality of fields are defined as one field in the second catalog definition information. be able to. For this reason, it is possible to forcibly adopt a processing method that is considered to be efficient for a predetermined processing request, and thus it is possible to select an optimal access method.

Further, when the index added in the first catalog information is defined in a part of one field, in the second catalog information, a part of the field is made one field, and the original Fields can be defined separately. For this reason, it is possible to forcibly adopt a processing method that is considered to be efficient, and thus it is possible to select an optimal access method.

Further, in the processing request, selection conditions for determining a match for each of a plurality of continuous fields defined in the first catalog information are specified, and the selection conditions are logical products. When combined, these fields can be defined as one field in the second catalog information. Therefore, a plurality of selection conditions described in the processing request can be changed to one selection condition, and an efficient processing method can be forcibly adopted. Thereby, there is an effect that an optimal access method can be selected.

Further, when the processing request describes a selection condition for judging a partial match between the value of one field defined in the first catalog information and the front or rear part, the second catalog information Can define one field as a plurality of fields. For this reason,
A processing request can be described not as a partial match for one field, but as a determination of a match for the entire field.
As a result, an efficient processing method can be forcibly adopted, so that an optimal access method can be selected.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration diagram of a database configuration information management system according to a first embodiment of the present invention.

FIG. 2 shows first and second embodiments according to the first embodiment of the present invention.
Of the first catalog information stored in the respective catalog information storage units.
FIG. 7 is a diagram showing an example of catalog information of the first and second catalogs.

FIG. 3 is a flowchart showing a processing procedure in the database configuration information management system of the present invention.

FIG. 4 is a flowchart showing the operation of a compiler in the database configuration information management system of the present invention.

FIG. 5 is a diagram showing an example of a make file according to the first embodiment of the present invention.

FIG. 6 is a diagram showing an example of a make file according to the first embodiment of the present invention.

FIG. 7 is a diagram showing a configuration diagram of a database configuration information management system according to a second embodiment of the present invention.

FIG. 8 is a diagram showing a configuration diagram of a database configuration information management system according to a third embodiment of the present invention.

FIG. 9 is a diagram showing an example of table definition information according to Embodiment 4 of the present invention.

FIG. 10 is a diagram showing an example of table definition information according to Embodiment 4 of the present invention.

FIG. 11 is a diagram showing an example of a make file according to Embodiment 4 of the present invention.

FIG. 12 is a diagram showing an example of a make file according to Embodiment 4 of the present invention.

FIG. 13 is a diagram showing an example of column definition information according to the fifth embodiment of the present invention.

FIG. 14 is a diagram showing an example of table definition information according to the sixth embodiment of the present invention.

FIG. 15 is a diagram showing an example of column definition information according to the seventh embodiment of the present invention.

FIG. 16 is a diagram showing an example of table definition information according to the eighth embodiment of the present invention.

FIG. 17 is a diagram showing an example of the configuration of a conventional database configuration information management system.

FIG. 18 is a diagram showing an example of column definition information in a conventional example.

[Explanation of symbols]

1 first catalog information storage unit, 1a catalog information to be changed, 2 second catalog information storage unit, 3 processing request analysis unit, 4 compiler, 5 computer, 6 input unit,
7, 8 processing request, 9, 10 access method, 11 column definition information, 12 index, 13 column definition information, 1
4 Index, 20 database storage unit, 21 database A, 22 database B, 23 catalog reference information, 36 second catalog information, 37 second catalog information, 38, 38a table definition information, 39 index, 4
0 One item, 41 index, 42 DATE, 43
Index, 44 one entry, 45 A1, 46 A2.

Claims (10)

[Claims] 1. A database storage unit for storing data, a first catalog information storage unit for storing first catalog information describing a configuration of data stored in the database storage unit, and the database storage unit An input unit for inputting a request for processing of data stored in the storage unit, and analyzing the processing request input from the input unit according to the first catalog information stored in the first catalog information storage unit. A database configuration information management system including a processing request analysis unit that obtains a predetermined access procedure for accessing a database, comprising: a second catalog information storage unit that stores second catalog information created for a predetermined processing request; In the input unit, the processing request analysis unit refers to the second catalog information stored in the second catalog information storage unit The processing request analysis unit stores the processing request input from the input unit in the second catalog information storage unit specified in the catalog reference information. A database configuration information management system, characterized in that a predetermined access procedure for accessing the database is obtained by analyzing according to second catalog information. 2. The second catalog information storage section changes the first catalog information stored in the first catalog information storage section into predetermined catalog information and stores the same as second catalog information. 2. The database configuration information management system according to claim 1, wherein: 3. The second catalog information storage unit changes a part of the first catalog information stored in the first catalog information storage, and stores the changed part of the information. 2. The database configuration information management system according to claim 1, wherein the database configuration information is stored as second catalog information. 4. The database configuration information management system according to claim 2, wherein said second catalog information storage unit stores catalog information in a predetermined file format. 5. The database configuration information management system according to claim 1, wherein the input unit inputs the processing request and the catalog reference information as one piece of information. 6. The first catalog information is composed of one or more fields, and the first catalog information storage unit stores an indexed field or an index for each field as the first catalog information. First index information indicating whether or not the field is absent is stored, and the second catalog information is composed of one or more fields.
The second catalog information storage unit stores, as the second catalog information, second index information indicating whether each of the fields is an indexed field or an unindexed field,
If the catalog information storage unit stores the first index information indicating that the predetermined field is an indexed field as the first catalog information, the second catalog information storage unit stores 2. The database configuration information management system according to claim 1, wherein said second index information indicating that said field is a field without an index is stored for a predetermined field. 7. The first catalog information storage section stores the first catalog information as the first catalog information for each of the fields.
And the first field name indicating the field is stored, and the second catalog information storage unit stores the second index information for each field as the second catalog information. And the second field name indicating the field is stored, and the first catalog information storage unit stores the first field name and the first index information for each of the two or more consecutive fields. Wherein the second catalog information storage unit stores a second field name and the second index information with the two or more consecutive fields as one field. 6. The database configuration information management system according to item 6. 8. The first catalog information storage section stores a first data length indicating a length of data stored in the field as the first catalog information, and stores one data stored in the field. A second index information storing unit that stores a first index length indicating that a part is to be indexed; and a second data indicating a length of data stored in the field as the second catalog information. And a second index length indicating that a part of the data stored in the field is to be indexed, wherein the first catalog information storage unit stores the first data length. When the first index length in which a part of data is indexed is stored, the second catalog information storage unit stores the first index length in the field having the first data length. 8. The database configuration information management system according to claim 7, wherein said second index length having one data length as an index is stored. 9. The first catalog information storage section stores a first field name indicating the field as the first catalog information, and the second catalog information storage section stores the second catalog information. When a second field name indicating the field is stored as information, and the first catalog information storage unit stores the first field name for each of two or more consecutive fields, the second catalog name is stored in the second catalog information storage unit. 7. The database configuration information management system according to claim 6, wherein the catalog information storage unit stores the second field name with the two or more consecutive fields as one field. 10. When the first catalog information storage section stores the first field name for at least one of the fields, the second catalog information storage section stores two or more of the one field. 10. The database configuration information management system according to claim 9, wherein two or more of said second field names are stored by dividing into two fields.