JPH05257765A - Database management system - Google Patents

Abstract

PURPOSE:To also manage a database in an old version by reproducing a database by purposes requested to be referred to according to the latest database by purposes extracted from a basic data base and differential data linked to this latest database. CONSTITUTION:When a database managing mechanism 14 generates a latest database 11 by purposes, a version managing mechanism 15 generates the data of difference between this database and the database by purposes at the time point of last extraction, registers the differential data on a database 12 for save and registers link relation between these registered differential data and the latest database 11 by purposes on a managing list 13. On the other hand, when the reproducing request of the database by purposes in the old version is executed through a terminal 2, first of all, the version managing mechanism 15 specifies the differential data of the database 12 for save to be linked to the database by purpose requested to be reproduced while referring to the managing list 13, continuously, the database by purposes requested to be reproduced is reproduced according to these specified differential data and outputted to the terminal 2 with the latest database 11 by purposes as a reference.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management system for managing a purpose-specific database extracted from a basic database, and more particularly to a database management system that enables management of an old edition of the purpose-specific database. ..

In recent years, SIS (strategic information system) has been introduced in companies. In this SIS, focusing on indicators of corporate activities that generate profits, it is specific data from a core database (for example, deposit and savings data of financial institutions) that has a characteristic that data is accumulated in a complicated manner according to purpose. A database is extracted, and business analysis such as product sales forecast is performed from various points of view using this purpose-based database. For example, if we explain using an example of statistical forecasts such as product sales forecasts, we will collect the data of the items to be analyzed for each specified period interval condition, and collect the collected data by month, season, half year, and year. After totaling, the optimum statistical analysis method is used based on this aggregated data to obtain forecast data for several years ahead, and this procedure is repeated in a certain cycle to constantly obtain the latest forecast data. ..

[0003]

2. Description of the Related Art In a conventional database management system,
The purpose-based database had a structure in which only the latest data was retained. That is, it does not have a function of saving the purpose-specific database collected for each generation for reuse.

Conventionally, therefore, the purpose-specific database extracted from the core database has been designed so as to have a totalized data from the beginning, paying attention to the items for business analysis.

[0005]

However, in such a conventional technique, in response to a request to make a new statistical prediction based on the data collected at a certain point in the past, the data at that point is collected again. There was a problem that we could not respond because we could not do it.

In order to deal with this, when it is necessary to save and reuse the collected purpose-specific database, the purpose-specific database is manually saved each time and is reused. It is possible to adopt the method of doing. However, if such a method is adopted, there is a problem in that the user's burden is heavy in terms of name management and version number management of the purpose-specific database, and also in terms of DASD space management. ..

The present invention has been made in view of the above circumstances, and is a database system having a structure in which a purpose-specific database is extracted from a basic database, and it is possible to manage an old version of the purpose-specific database. The purpose is to provide a new database management system.

[0008]

FIG. 1 illustrates the principle configuration of the present invention. In the figure, 1 is a database system including the present invention, and 2 is a terminal included in the database system 1.

The database system 1 includes a backbone database 10, a latest purpose-specific database 11, a backup database 12, a management book 13, a database management mechanism 14, and a version number management mechanism 15.

The basic database 10 manages basic data accumulated over time. The latest purpose-specific database 11 manages the latest purpose-specific data extracted from the core database 10. The evacuation database 12 manages the difference data between the latest purpose-specific database 11 and the purpose-specific database at the time of the previous extraction, and, in addition to this, the purpose-specific database at each extraction time and the extraction time Manages the difference data from the purpose-specific database at the time of the previous extraction.

The management book 13 is the latest purpose-specific database 1
The connection relationship between 1 and the difference data of the save database 12 connected to this is managed. Database management mechanism 1
Reference numeral 4 executes database management such as update processing of management data of the backbone database 10 and generation processing of the latest purpose-specific database 11. The version number management mechanism 15 is, for example,
It is expanded to the database management mechanism 14 to execute the generation process of the difference data of the save database 12 and the reproduction process of the old-edition purpose-specific database.

[0012]

In the present invention, when the database management mechanism 14 generates the latest purpose-specific database 11 at the end of the month, for example, the version number management mechanism 15 creates the latest purpose-specific database 11 and the purpose-specific database at the time of the previous extraction. The differential data of 1 is generated and registered in the evacuation database 12, and the connection relation between the registered differential data and the latest purpose-specific database 11 is registered in the management book 13. According to the registration process of the version number management mechanism 15, the management book 13 manages the connection relation between the latest purpose-specific database 11 and the difference data of the save database 12 connected to this. At this time, the purpose-based database at the time of the previous extraction, which has been managed as the latest version, is deleted as usual.

On the other hand, when there is a reproduction request for the old version of the purpose-specific database via the terminal 2, the version number management mechanism 15 first refers to the management book 13 to create the purpose-specific database for the reproduction request. Evacuation database 12 to be linked
Of the purpose-specific database at the time of each extraction in sequence according to the specified difference data based on the latest purpose-specific database 11 as a reference. It reproduces and outputs to the terminal 2.

As described above, according to the present invention, in the database system 1 having the structure in which the latest latest purpose-specific database 11 is extracted from the basic database 10, the old-version purpose-specific database can be managed. Become. From now on, it becomes possible to meet a request such as a new statistical prediction based on the purpose-based database collected at a certain point in the past. Moreover, it becomes possible to manage the old purpose-specific database with a small memory capacity by using the difference data.

[0015]

EXAMPLES The present invention will be described in detail below with reference to examples. FIG. 2 shows a database system 1 including the present invention.
1 illustrates an example of the system configuration of FIG. In the figure, the same components as those described in FIG. 1 are indicated by the same symbols.

Reference numeral 20 is an operating system, 21 is a TSS for executing interactive processing with the user, 22 is a VTAM for controlling communication control processing with the terminal 2, and 23 is a database operation management newly provided to support the data management mechanism 14. Assistance mechanism, 24 is the backbone database 1 described in FIG.
It is a relational database that expands 0, etc. In this embodiment, the version number management mechanism 15 is configured to be deployed in the database operation management support mechanism 23.

The table data constituting the purpose-specific database is extracted from the core database 10 expanded in the relational database 24 at every elapse of a certain period, and is changed / added / deleted with the lapse of time. According to the basic data of the basic database 10 that will be different, data different from the table data extracted last time will be managed.

FIG. 3 shows an example of updating the table data which constitutes this purpose-specific database. Like this, 5
The latest table data extracted as of the 20th of the month is 4
Part of the management data of the table data extracted as of the 20th of the month is changed, new management data is added, and part of the management data is deleted (there is nothing deleted in this example). It will be generated.

As described with reference to FIG. 1, the database system 1 of the present invention manages not only the latest extracted table data but also the difference data between the latest table data and the previously extracted table data. By doing so, it is possible to reproduce the table data extracted last time. Here, when there is a possibility that a reproduction request for the table data extracted earlier is issued, the table data at the extraction time point and the table data at the previous extraction time point with respect to the extraction time point are extracted at each extraction time point. The difference data will be managed.

This difference data is generated by, for example, sorting the latest table data in ascending order of date / time key items (displaying the date / time of data registration) and sorting the previous table data in ascending order of date / time key items. Is executed by comparing the sorted two table data item by item. The difference data thus generated is stored in the evacuation database 12 described in FIG. 1, which is expanded in the relational database 24 according to evacuation format data having a data structure as shown in FIG. 4A, for example. Will be.

Here, in the data structure shown in FIG. 4 (a), the "change type" indicates which change type among the changes, additions and deletions, the difference between the two data is. For example, "U" is registered for change, "I" is added for addition, and "D" is registered for deletion.
The “location information” indicates in which table of which schema the latest table data has been expanded. “Line number” displays the line number of the data having the difference. The “item name” displays the item name with a difference, and when the change type is addition, the NULL value is registered. "Item data" displays the data value of the item that has a difference. If the change type is changed or deleted, the data value of the previous table data is registered, and if the change type is added, A NULL value will be registered.

FIG. 4 (b) shows the latest table data extracted on May 20 shown in FIG. 3 and April 2 shown in FIG.
The save format data which is the difference data with the table data extracted on the 0th day is illustrated. In this save format data example,
It is assumed that “C1” in the table data shown in FIG. 3 represents a date / time key item, and the table data of May 20 / April 20 shown in FIG. It is assumed that they are sorted in ascending order.

The management book 13 provided in the database system 1 of the present invention executes the above-mentioned save format data management processing, as well as management processing for enabling reproduction of the table data extracted previously.

FIG. 5 shows an embodiment of the management data of the management book 13. As shown in this embodiment, the management book 1
Reference numeral 3 manages table definition information, schema definition information, relationship information, and version number information.

Explaining these management information, the table definition information includes the latest table in addition to managing the schema name / table name of the basic database 10 to be expanded, its creation date, and creator ID. The schema name / table name where the data is expanded, the creation date of the table data, and the creator ID are managed, and the schema name where the save table data that expands the save format data is expanded is managed as shown in the figure. / Manages the table name, the creation date of the table data, and the creator ID. Here, since the save table data is created by the version management mechanism 15, the expanded schema is automatically generated by the version management mechanism 15.

The schema definition information includes the schema name of the generated schema, the creator type of the schema, the creation date,
Manage the creator ID. The schema definition information pointed to in the figure represents the definition information regarding the schema storing the save format data. The relationship information manages a parent-child relationship between the table data that is a parent and the table data that is a child, and the version type names that are treated as being homogeneous according to the parent-child relationship. That is, as shown in FIG. 6, the latest table data is a child, and the save table data that is the difference data between this latest table data and the previous table data is the parent. It manages the parent-child relationship of linked save table data. The version number information manages the number of generations to save and the save schema name of the version number type name managed by the relationship information.

The version number management mechanism 15 uses the management function of the management book 13 to generate save format data and stores it in the save database 12, and also uses the saved save format data to save the previous version. Processing for reproducing table data, which is a purpose-specific database, will be executed.

FIG. 7 shows an embodiment of a processing flow executed when the version number management mechanism 15 generates save format data. In FIG. 8, it is executed when the version number management mechanism 15 reproduces the previous table data. An example of the processing flow to do is shown in figure.
Next, the database management processing of the present invention will be described in detail according to these processing flows.

The version number management mechanism 15 is, for example, 2 in the month.
When it is time to update the latest table data of the purpose-specific database by the 0th day, as shown in the processing flow of FIG. 7, first, in step 1, according to the management data of the management book 13, the backbone database The table T1 of the schema S1 of the database 1 that is the expansion destination of 10 is specified, and the table data of the purpose-specific database is extracted by performing a conditional search from the specified expansion destination, and the extracted table data is date / time key item. Execute the process to sort and expand in memory. In this way, as shown in FIG. 9, the table data of the second version, which is the latest table data of the purpose-based database, is expanded on the memory.

Next, in step 2, according to the management data of the management book 13, the table data of the purpose-specific database extracted at the time of the previous extraction (the latest table data from the time of the previous extraction to this time) The table T1 of the schema S2 of the database 2 which is the deployment destination of (the existing one) is specified, and the table data of the purpose-specific database is read from the specified deployment destination,
The read table data is sorted by the date / time key item and expanded on the memory. In this way, as shown in FIG. 9, the table data of the first version, which is the table data of the old version of the purpose-specific database, is expanded on the memory.

Then, in step 3, the variable i is set to "1", and in the following step 4, the second value expanded on the memory is set.
The i row data of the plate table data is taken out, and in the following step 5, it is judged whether or not the i row data was taken out in step 4.

When it is determined in step 5 that the i-row data has been extracted, the process proceeds to step 6, and it is determined whether or not the i-row data extracted in step 4 exists in the table data of the first version. If it is determined that there is not, "I" is registered in the "change type" of the save format data described in FIG. 4A, and NULL values are registered in the "item name" and "item data"("linenumber"). / Register corresponding values for "location information"). On the other hand, when the i-line data extracted in step 4 has a different value in the table data of the first version, "U" is registered in "change type" and an item name having different item data in "item name" Further, the item data of the item name of the table data of the first edition is registered in the “item data” (corresponding values are also registered for “line number” / “location information”). On the other hand, if there is line data that exists only in the first edition table data, "D" is registered in the "change type" and the item data of the first edition table data is registered in the "item data". (Corresponding values are also registered for "item name" / "line number" / "location information").

Then, in step 7, the value of the variable i is incremented by 1, and then the process returns to step 4. By repeating the processing of step 4 to step 7 in this way, as shown in FIG. 9, the save format which is the difference data between the table data of the latest purpose-specific database and the table data extracted last time. The data will be expanded on the memory.

On the other hand, when it is determined in step 5 that the i-th row data cannot be extracted, it means that the generation of the save format data is completed, so the process proceeds to step 8, and as shown in FIG. In the table T1 of the schema S2 of the database 2 prepared to store the latest table data, the second version table data developed in the memory is stored as a replacement for the first version table data. Then, in the following step 9, FIG.
As shown in, the schema SY of the backup database 12 prepared for storing the backup format data
The save format data (first version) expanded on the memory is stored in the table T1 of the SSCM1, and in the following step 10, as shown in FIG. In accordance with this, the table definition information of the management book 13 is updated, and the parent-child relationship of the relationship information is updated.

In this way, the version management mechanism 15 executes the processing flow shown in FIG. 7, whereby the latest table data of the purpose-specific database and the difference between the latest table data and the previously extracted table data are obtained. The data will be processed so as to manage the save format data.

Next, the processing flow shown in FIG. 8 will be described. When there is a request to refer to the table data of the purpose-specific database previously extracted via the terminal 2, the version number management mechanism 15 first, as shown in the processing flow of FIG.
In step 1, the table T1 of the schema S2 of the database 2, which is the storage location information of the latest table data, is obtained from the management data of the management book 13, and the storage location of the save format data having a parent-child relationship with the latest table data. Schema SYSS of backup database 12 which is information
Get the table T1 of CM1.

Next, in step 2, the latest table data is read from the expansion destination obtained in step 1, and the read table data is sorted by the date / time key item and expanded on the memory. In this way, FIG.
As indicated by 0, if explained at the end stage of the processing flow of FIG. 7, the table data of the second version stored in the table T1 of the schema S2 of the database 2 will be expanded on the memory.

Subsequently, in step 3, the save format data having a parent-child relationship is read from the expansion destination obtained in step 1 and expanded on the memory. In this way, as shown in FIG. 10, if explained at the end stage of the processing flow of FIG. 7, the save data of the first version stored in the table T1 of the schema SYSSCM1 of the save database 12 is stored in the memory. Will be deployed on top.

Subsequently, in step 4, "1" is set to the variable i, in step 5, the i row data of the save format data expanded on the memory is taken out, and then in step 6
Then, it is determined in step 5 whether or not the i-th row data can be extracted.

When it is determined in step 6 that the i-line data has been extracted, the process proceeds to step 7, where the "change type" of the i-line data extracted in step 5 is checked and this "change type" is added. Is displayed, the line data indicated by the "line number" is deleted from the second version table data expanded in the memory. on the other hand,
When this "change type" is displaying "U" indicating a change, it is indicated by the "item name" in the line data indicated by the "line number" of the table data of the second version loaded in the memory. Change the item name item data to "item data". On the other hand, when this "change type" is displaying "D" indicating deletion, the line data having the item data "item data" at the "line number" position of the table data of the second version loaded in the memory Is newly inserted.

Then, in step 8, the value of the variable i is incremented by 1, and then the process returns to step 5. By repeating the processing of steps 5 to 8 in this manner, the first version table data extracted last time is expanded on the memory as shown in FIG.

In this way, the version management mechanism 15 executes the processing flow of FIG. 8 to perform processing so as to reproduce the table data of the purpose-specific database previously extracted.

Finally, the overall flow of the database management process of the present invention will be described with reference to FIG. Here, FIG.
In 1, the thick line indicates the data flow, the thin line indicates the instruction,
The following or corresponds to or in the figure.

Version number definition First, the user specifies the version number type name (name given to a group of homogeneous tables under version number management), the number of generations of the table to be held, and the schema name to be saved. Register in management book 13.

First Version Table Data Creation Next, the user lists the table definition information for the backbone database 10, and selects a specific table T1 there. Then, the conditions for data extraction up to April 20, 1991 are specified, and T is used as the table name for the first edition.
Specify 1A. In response to the designation of this extraction condition, the version number management mechanism 15 creates the table T1A which is the table data of the first version of April 20, 1991.

Creation of Second Edition Table Data Next, the user lists the table definition information for the backbone database 10, where the specific table T1 is specified.
Select. Then, the conditions for data extraction up to May 20, 1991 are specified, and the table T1A of the first version and the version number type name A are specified to declare that the version number management is started. In response to this declaration, the version management mechanism 15
Table T, which is the table data for the second edition of the 20th edition of the month
1A is created, save format data is generated, and the parent-child relationship between the second version table T1A and this save format data is registered in the management book 13.

List of Version Numbers Next, the user lists the table definition information and selects a specific table T1 there. Then, the version number type name A is specified according to the version number definition information to instruct the execution of the version number list function. In response to this instruction, the version management mechanism 15
Table T of April 20, 1991 edition and May 20, 1991 edition
Display 1A table definition information.

Retrieval of Old Version Table Subsequently, the user gives an instruction to retrieve the table T1A of the April 20, 1991 edition of the first edition from the list of edition numbers. In response to this instruction, the version number management mechanism 15 is the second version 91.
The table T1A of April 20, 1991, which is the first version, is reproduced and output from the table T1A of May 20, 2014 and the save format data having a parent-child relationship with the table T1A.

Deletion of Old Version Table Subsequently, the user gives an instruction to delete the table T1A of the first version, April 20, 1991, using the version number list. In response to this instruction, the version number managing mechanism 15 deletes the table T1A of the April 20, 1991 version of the first version by deleting the save format data having a parent-child relationship.

[0050]

As described above, according to the present invention,
The database system has a structure in which the latest purpose-specific database is extracted from the core database, and it becomes possible to manage the old-purpose purpose-specific database. From now on, it becomes possible to meet a request such as a new statistical prediction based on the purpose-based database collected at a certain point in the past. Moreover, it becomes possible to manage the old purpose-specific database with a small memory capacity by using the difference data.

[Brief description of drawings]

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

FIG. 2 is an embodiment of the system configuration of the database system of the present invention.

FIG. 3 is an explanatory diagram of an example of updating table data.

FIG. 4 is an explanatory diagram of save format data.

FIG. 5 is an example of management data of a management book.

FIG. 6 is an explanatory diagram of related information.

FIG. 7 is an example of a processing flow executed by a version number management mechanism.

FIG. 8 is an example of a processing flow executed by a version number management mechanism.

FIG. 9 is an explanatory diagram of a process of the embodiment.

FIG. 10 is an explanatory diagram of a process of the embodiment.

FIG. 11 is an explanatory diagram of the processing of the present invention.

[Explanation of symbols]

 1 Database System 2 Terminal 10 Core Database 11 Latest Purpose Database 12 Evacuation Database 13 Management Book 14 Database Management Facility 15 Version Management Facility

Claims (1)

[Claims] 1. A database management system for managing a purpose-specific database extracted from a core database, which manages the latest purpose-specific database to be extracted, and uses this latest purpose-specific database as a reference at each extraction time point. And adopting a configuration for managing the difference data between the purpose-specific database at the time of extraction and the purpose-specific database at the time of the previous extraction with respect to the time of extraction, and one or more of the above-mentioned differences linked to the same latest purpose-specific database. When there is a reference request for a purpose-specific database extracted in the past by adopting a configuration for managing the data connection relationship, the reference request is issued according to the latest purpose-specific database and the difference data connected to the latest purpose-specific database. Characterized by processing to reproduce the purpose-based database Database management systems.