JP2000181769A - Redundant database device - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To make it possible to equalize a duplicated database (DB) even when data is updated or when data includes a reference to another data. SOLUTION: The generation time and update time of the data are recorded together with the contents of the data, and D
When data is copied from B1 to DB2 on the slave side, the data identifier of the data to be copied is recorded in DB2 on the slave side together with the contents of the data, and the valid data extracting unit 4 designates a user from DB2 on the slave side. Data generated or updated after the time is generated or updated from the master side DB1 before the specified time, and data not yet extracted from the slave side DB2 is extracted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a duplicated database device in which a master database and a slave database are duplicated, and more particularly to an equivalent processing of the master database and the slave database.

[0002]

2. Description of the Related Art FIG.
FIG. 8 is an explanatory diagram showing a database equivalence process in the conventional duplex database device shown in Japanese Patent Publication No. 8; In the figure, reference numerals 1 and 2 denote databases in which one is a master side and the other is a slave side. Reference numeral 11 denotes a data storage unit for storing collected data in the database 1; 12, a master / slave information storage unit for storing master / slave information; and 21, a collected data in the database 2. The data storage unit 22 is also the master /
This is a master / slave information storage unit in which slave information is stored.

Next, the operation will be described. The collected data is stored in the data storage units 11 and 12 of each of the databases 1 and 2.
1 are stored in the respective databases 1 and 2
The master / slave information indicating whether each is a master or a slave is stored in each master / slave information storage unit 1.
Record according to 2 or 22. In FIG. 6,
An example of recording in the case where the database 1 is a master and the database 2 is a slave is shown. When the collected data # 3 and # 4 were recorded, the database 1 was stopped. Therefore, the collected data is recorded only in the data storage unit 21 of the database 2, and the master / slave information storage unit 22 stores "master". Is recorded. Equalization of these two databases 1 and 2 is performed by reading out only the data when the databases 1 and 2 are masters.

[0004] Therefore, the operations of the databases 1 and 2 that can be handled in this database equivalence processing are only addition of collected data, and cannot handle updating of existing data. For example, when the collected data # 7 is updated while the database 1 is stopped, the master / slave information in the master / slave information storage units 12 and 22 becomes “master” in both the databases 1 and 2, and which data is It is indistinguishable from correct. In addition, the collected data stored in the data storage units 11 and 21 includes a data identifier indicating a reference to another collected data, for example, a tuple ID in a relational database, an object ID in an object database, and the like, and a reference relationship exists. If the data straddles the database 1 on the master side and the database 2 on the slave side, the data reference relationship will be incorrect.

[0005]

Since the conventional redundant database apparatus is configured as described above, the collected data can be added, but the existing collected data cannot be updated. Further, when collected data having a reference relationship extends between the database 1 on the master side and the database 2 on the slave side, there is a problem that the reference relationship of the data becomes incorrect.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems. Even when data is updated or data includes a reference to another data, the database is constructed in a dual system. It is an object of the present invention to obtain a duplicated database device that can equalize the data.

[0007]

According to the present invention, there is provided a duplex database device for recording, for each data stored in each database, the contents of the data and the generation time and update time of the data. is there.

The duplicated database device according to the present invention records the data identifier of the copied data together with the contents of the data in the slave database when copying the data from the master database to the slave database. It was made.

[0009] The duplicated database device according to the present invention is provided with a valid data extracting unit, and when equalizing the databases of both systems, from the slave database,
The data whose generation time or update time is later than the time specified by the user is extracted, and the data whose generation time or update time is earlier than the time specified by the user is still extracted from the database on the slave side. This is to retrieve data that is not stored.

[0010] The duplicated database device according to the present invention is provided with a data reference conversion unit, and in the data retrieved by the valid data retrieval unit, data having a reference relationship extends between the master database and the slave database. When there is a data, the reference destination of the data copied from the master database to the slave database is converted using a data identifier in the master database.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below. Embodiment 1 FIG. FIG. 1 is a configuration diagram showing a part of the duplex database apparatus according to the first embodiment of the present invention that performs database equivalence processing. In the figure, reference numeral 1 denotes a database on the master side where a database equivalence process is performed, and reference numeral 2 denotes a database on the slave side. Reference numeral 3 is created by extracting data from the database 1 on the master side and the database 2 on the slave side based on the time specified by the user, and connecting the databases 1 and 2 of both systems at the specified time. This is the database after linking.

When the master database 1 and the slave database 2 are equalized, the slave database 2 outputs a time later than the time specified by the user as a generation time or an update time. The master database 1 is retrieved and the generation time or the update time is earlier than the time specified by the user, and is copied from the master database 1 to the slave database 2. ,
This is a valid data extracting unit that extracts data that has not been extracted from the database 2 on the slave side.

Reference numeral 5 denotes a database 3 after connection with the data retrieved by the valid data retrieval section 4.
Is a data identifier conversion unit that converts the data identifier of the data so that the data identifier is unique. Reference numeral 6 denotes a data identifier correspondence table for storing the correspondence between the data identifier before the change and the data identifier after the change with respect to the data identifier converted by the data identifier converter 5. Reference numeral 7 denotes a data reference conversion unit that converts, based on the data identifier correspondence table 6, a reference to other data included in the data so as to be a correct reference in the database 3 after connection.

Next, the operation will be described. Here, FIG.
FIG. 3 is an explanatory diagram showing the format of data used in the duplex database apparatus according to the first embodiment. FIG. 4A shows an example of data stored at the time of data generation or change in the databases 1 and 2 of both systems. FIG. 2B is an example of data stored at the time of data copying. When data is generated or changed in each database 1 or 2, as shown in FIG. 2A, the time when the data is generated and the time when the data is updated are recorded in the data together with the contents. You. It is assumed that (A, 3) is given to this data as a data identifier. FIG. 2B shows a copy of the data whose data identifier is (A, 3) (hereinafter, referred to as data (A, 3). The same applies to data of other data identifiers). As described above, the data identifier (A, 3) of the data (A, 3) as the copy source is also recorded in the data. Here, it is assumed that the data is given a data identifier (A, 5). If this data (A, 5) is not a copy,
The copy source item is blank (n / a).

Next, the retrieval of the data recorded as described above will be described. Here, FIG. 3 is an explanatory diagram showing a flow of extracting data in the valid data extracting section 4. In this case, in the database 1 on the master side,
Generation of data (A), update of data (A), generation of data (B), update of data (B), generation of data (C), update of data (C), update of data (B), data The generation of (D) and the update of data (D) are performed in order. In the database 2 on the slave side, the data (B) is copied from the database 1 on the master side, and the data (C) from the database 1 on the master side is copied. , Update of data (B), generation of data (D), update of data (D), and update of data (C) are performed in order. Here, assuming that the time shown as the connection plane in FIG. 3 is the time designated by the user, data (A) is transmitted from the database 1 on the master side, data (B) and data are transmitted from the database 2 on the slave side. (C) and data (D) are respectively extracted as the latest data.

The data is taken out according to the following procedure. First, database 2 on the slave side
Are extracted from the data included in the data of the update time later than the time specified by the user, that is, those after the time of the connection plane. In addition, as for the data in which the update time is not recorded, the data whose generation time is after the time of the connection plane is extracted.
Next, from the data contained in the database 1 on the master side, the update time is before the time specified by the user, that is, before the time of the connection plane, and is the data copied to the database 2 on the slave side, Those that have not been retrieved from the database 2 on the slave side are retrieved. Also in this case, as for the data for which the update time is not recorded, the data whose generation time is not taken out is taken out from the data before the time of the connection plane. Whether the data copied to the slave-side database 2 has not been retrieved by the slave-side database 2 can be determined by determining whether the data identifier of the data to be retrieved by the master-side database 1 is included in the copy source information in the slave-side database 2.
It is determined whether or not the data has been extracted.

Hereinafter, the above procedure will be described with reference to the example shown in FIG. First, database 2 on the slave side
When the update times of all the data existing in the slave side, that is, the data (B), the data (C), and the data (D) are examined, they are all after the time designated as the connection plane. , Data (B), data (C), and data (D) are extracted.
Next, when the update times of all the data existing in the database 1 on the master side, that is, the data (A), the data (B), the data (C), and the data (D) are checked, the update times before the time designated as the connection plane are found. Are data (A) and data (C). Further, when the data (A) and the data (C) are copied to the slave database 2 and those already retrieved from the slave database 2 are examined, the data (C) corresponds to the data (C). Therefore, only the data (A) is extracted from the database 1 on the master side. In this way, data (A), data (B), data (C), and data (D) are extracted from the linked database 3 without duplication.

Next, generation of the data identifier correspondence table 6 by the data identifier conversion unit 5 and conversion of data reference by the data reference conversion unit 7 will be described. 4 is an explanatory diagram showing the flow of the data reference conversion, and FIG. 5 is an explanatory diagram showing an example of the data identifier correspondence table 6. here,
As shown in FIG. 4, from the database 1 on the master side, data (A, 4) and data (B,
6), the data (A, 6) are taken out, and the data (B,
8), data (A, 6), and data (B, 9) are taken out to constitute the linked database 3.
In the database 1 on the master side, the data (B,
Reference to 6) is to data (A, 4) and data (B, 7)
Is included in the data (A, 6), and in the database 2 on the slave side, the reference to the data (B, 8) is the data (A, 3), and the reference to the data (B, 9) is the data. (A, 6). The data (A, 6) of the master database 1 is copied to the slave database 2 as data (A, 3), and the data (B, 7) of the master database 1 is copied to the slave database 2. It is assumed that the data is copied as data (B, 8).

When reconstructing the data retrieved by the valid data retrieval section 4 into the linked database 3, the data identifiers need to be replaced so that there is no duplication. Here, FIG. 5 shows the contents of the data identifier correspondence table 6 created with the replacement of the data identifier in the example shown in FIG.

In the example shown in FIG. 4, the data (A, 4) of the database 1 on the master side includes a reference to the data (B, 6), but corresponds to the data (A, 4). The reference in the data (A, 1) of the linked database 3 needs to be changed to the data (B, 1) of the linked database 3 corresponding to the data (B, 6). When two data having a reference relationship are extracted across the database 1 on the master side and the database 2 on the slave side, the data (A, 6) is read from the data (A, 6) of the database 1 on the master side. The reference to 7) is converted from the data (A, 2) to the reference to data (B, 2) corresponding to the data (B, 8) extracted from the database 2 on the slave side in the database 3 after connection. Need to be done.

In this case, the data identifier conversion of the reference data is performed according to the following procedure. First, the data identifier correspondence table 6 is used to check whether both the data for changing the reference destination data and the reference destination data have been extracted from the database 1 on the master side or the database 2 on the slave side. As a result, if the data for changing the reference data and the data of the reference data are both extracted from the database 1 on the master side or the database 2 on the slave side, the data after the concatenation of the reference data is obtained. The data identifier in the database 3 is obtained from the data identifier correspondence table 6 and is used as the converted data identifier.
If one of the data for changing the reference data and the data of the reference data is retrieved from the database 1 on the master side and the other is retrieved from the database 2 on the slave side, the data copied from the reference data is obtained. Is obtained from the copy source data identifier stored in the data, and the data identifier in the database 3 after the linking of the data is obtained from the data identifier correspondence table 6 to be the converted data identifier.

Hereinafter, the above procedure will be described with reference to the examples shown in FIGS. Master side data (A,
Since the reference from 4) to the data (B, 6) on the master side is both taken out from the database 1 on the master side, the data identifier in the database 3 after linking the reference destination data is stored in the data identifier correspondence table. 6 to obtain the converted data identifier. In other words, it can be seen from the data identifier correspondence table 6 shown in FIG. 5 that the data (B, 6) on the master side has been changed to the data (B, 1) in the database 3 after connection, so the data (A, , 4), that is, the reference in the data (A, 1) of the database 3 after linking is defined as data (B, 1).

The reference from the data (A, 6) on the master side to the data (B, 7) on the master side is performed by extracting the data (B, 7) on the master side into the database 3 after connection. Instead, the slave side data (B,
8), the data copied from the reference destination data is obtained from the copy source data identifier stored in the data, and the data identifier in the database 3 after the concatenation of the data is replaced with the data identifier The converted data identifier is obtained from the correspondence table 6. That is, the data (B, B) referenced from the data (A, 6) on the master side.
Search for data having 7) as the copy source data identifier from the data extracted from the database 2 on the slave side. As a result, the data (B,
8) is found. Then, the data (B,
8) The data identifier in the database 3 after the connection is
Since the data (B, 2) is obtained by referring to the data identifier correspondence table 6, the data (A, 6) on the master side, that is,
The reference in the data (A, 2) of the database 3 after the connection is defined as data (B, 2).

In this manner, the data reference conversion unit 7 has the data reference relation extracted from the master database 1 and the slave database 2 in the data extracted by the valid data extraction unit 4. When there is a data straddling, the data copied from the reference data is obtained from the copy source data identifier stored in the data, and the data identifier of the data is obtained from the data identifier correspondence table 6. By using the converted data identifier as the data identifier, the database 3 after linking creates the database 3 after linking while properly maintaining the reference relation.

As described above, according to the first embodiment,
By recording the data generation time and update time together with the data content, it is possible to determine which data is the latest data in both systems not only for data generation but also for update, Also, even when data is copied from the master database 1 to the slave database 2, it is possible to identify data stored separately in both systems, and furthermore, data generation and updating are performed. In the database, the latest data can be extracted from the databases 1 and 2 constructed by both systems without any duplication, and the database 3 after the connection can be constructed. Therefore, even if the data is updated, the master can be obtained. The correct data can be extracted from the slave and slave sides without duplication, and the linked database 3 can be constructed. Even if data that is fetched across the database 1 on the master side and the database 2 on the slave side, by converting the data identifier representing the reference into the database 3 after linking, a correct reference relationship can be established in the database 3 after linking. Even if the data includes a reference to another data and the reference extends to the master database 1 and the slave database 2, the reference relationship should be correctly maintained in the linked database 3. And the like.

[0026]

As described above, according to the present invention, the generation time and the update time of the data stored in each database are recorded together with the contents of the data. In the case of data update, it is possible to determine which data is the latest data in the data of both systems, so even if the data is updated, the master database and the slave Correct data can be extracted from both databases without duplication, and the duplicated database can be equalized, so that there is an effect that a duplicated database device that can flexibly operate the system can be obtained.

According to the present invention, when data is copied from the master database to the slave database, the data identifier of the copied data is recorded in the slave database together with the contents of the data. When data is copied from the master database to the slave database, data stored separately in both systems can be identified, so that both the master database and the slave database can be identified. Thus, there is an effect that correct data can be extracted from the database without duplication, and it is possible to equalize a duplicated database.

According to the present invention, data whose generation time or update time is later than the time specified by the user is extracted from the database on the slave side, and data whose generation time or update time is specified by the user is specified from the database on the master side. At the time before the time, the data that has not been retrieved from the slave side database is configured to be retrieved. Since the database can be constructed after being taken out and connected, there is an effect that the duplicated database can be equalized.

According to the present invention, if there is any data having a reference relationship between the master database and the slave database in the extracted data, the slave database is replaced with the master database. The reference destination of the data copied to the database of the master is converted using the data identifier in the database on the master side. If the data identifier that represents the reference is converted to the linked database even if it spans the database of the slave and the database on the slave side, the linked database can maintain the correct reference relationship. There is an effect that it is possible to equalize the database.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a duplicated database device according to Embodiment 1 of the present invention.

FIG. 2 is an explanatory diagram showing a data format according to the first embodiment.

FIG. 3 is an explanatory diagram showing a flow of extracting data according to the first embodiment;

FIG. 4 is an explanatory diagram showing a flow of data reference conversion according to the first embodiment;

FIG. 5 is an explanatory diagram showing an example of a data identifier correspondence table in the first embodiment.

FIG. 6 is an explanatory diagram showing a database equivalent process in a conventional duplex database device.

[Explanation of symbols]

1 Master side database, 2 Slave side database, 3 Connected database, 4 Effective data retrieval section, 7 Data reference conversion section.

Claims (4)

[Claims] 1. Data is extracted from a duplicated master database and slave database based on a time specified by a user, and the master database and the slave database are connected at the specified time. In the dual database device that creates a database after linking, when the data is generated or updated, the generation time and update time of the data are stored for each data stored in the database. A duplicate database device characterized by recording with the contents. 2. Data is extracted from the duplicated master database and slave database based on a time specified by a user, and the master database and the slave database are connected at the specified time. A data identifier of data to be copied stored in the master-side database when copying data from the master-side database to the slave-side database. In the slave-side database together with the contents of the data. 3. Extracting data from the duplicated master database and slave database based on the time specified by the user, and connecting the master database and the slave database at the specified time. In the dual database device that creates a database after linking, when generating or updating the data, the generation time or update time of the data is recorded together with the content of the data, and the slave from the master-side database is recorded. When copying data to the database on the side, the data identifier of the copied data in the database on the master side is recorded in the database on the slave side together with the content of the data, and the database on the master side and the data on the slave side are recorded. When performing equalization with the database, From the database on the slave side, data having a generation time or an update time later than the time specified by the user is extracted, and from the database on the master side, a time earlier than the time specified by the user is generated. A duplicated database apparatus, comprising: a valid data extracting unit that retrieves data that has a time or an update time and is copied to the database on the slave side and has not been extracted from the database on the slave side. 4. When concatenating the data retrieved by the valid data retrieval unit to create a coupled database, the retrieved data has a data reference relationship with the data retrieved from the master database. If there is data spanning between data retrieved from the slave database, the data identifier in the master database of data copied from the master database and recorded in the slave database is used. 4. The dual database apparatus according to claim 3, further comprising a data reference conversion unit for converting a reference destination of the data.