JP2000057046A - Method, device and system for managing data and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data managing device capable of preserving and restoring data for each specified data unit. SOLUTION: Based on serial information preset by a serial information holding part 14, a preserving processing part 12 extracts a specified object set to become a preserving object from a database(DB) 16 through a DB control part 11, exports it to an external file 17 and preserves it. In cooperation with the DB control part 11 and an object ID(OID) managing part 15, a restoring processing part 13 imports the object set preserved in the external file 17 to the DB 16, reflects it and restores the object set at the time point of preservation. In order to maintain consistency in the restoring processing of the object set, the OID managing part 16 manages the former OID of the object at the time point of preserving in the external file 17 and a new OID at the time point of restoring the DB 16 while making them correspondent.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a database management technique, for example, a data management method in an object relational database.

[0002]

2. Description of the Related Art An object-oriented database management system (Object-Oriented Data Base Managemn) which integrally encapsulates data and algorithms for realizing a database based on an object-oriented concept.
et System (hereinafter, OODBMS) is known. This type of OODBMS is effective for atypical records such as multimedia information such as image and audio data, and a set of objects is defined as an object class for each object of the same type. Defines an internal state variable, a response to a message, and the like. Also, a hierarchical structure is defined for each class, and it is possible to construct a database having a structural expression by modeling according to a real complex event.

On the other hand, a relational database management system (Relation) which occupies the mainstream in the current database.
al Data Base Management System (RDBMS)
Object relational database management system (Object Relational Data Base Manageme), which is an object-oriented extension that adds flexibility to data manipulation and object-oriented applications.
nt System (hereinafter, ORDBMS). This ORD
In the BMS, for example, it is possible to reconstruct the existing RDBMS in an object-oriented manner while keeping the assets of the RDBMS alive.

In the ORDBMS, data in a database is called an object, and each object belongs to one class, and the class is implemented as a “table” on the database. Thus, conceptually, a "table" can be thought of as a container containing objects of a particular class.

In the ORDBMS, an identifier called OID (Object ID) is assigned to every object in the database. This OID is used when the object is inserted into the database,
It is given by S. The OID is used to identify an object, that is, uniquely identify an object in a database. Objects having the same OID are regarded as the same object. Also, the OID used once in object identification
Is not reused.

[0006] The OID is also used to express a reference relationship between objects from a specific object to another object. For example, an object a
When the object a1 refers to the object b1 for 1 and b1, 1 in the object a1
It is configured to have the OID of the object b1 in one attribute.

Further, for example, a database once built on a physical disk is stored in a file for storing objects in the database for reasons such as data maintenance, disk replacement, system migration, and the like. A technique for restoring the image to a computer has been conventionally known. In this technique, processing called export and import is generally performed. Specifically, the export is a process of saving a part or all of the objects of a database in an external file or the like in table units. On the other hand, import is a process of restoring a corresponding object in a database by reading a file written and saved by export.

FIG. 8 is a functional block diagram of a database management system showing a method of saving and restoring objects in a conventional ORDBMS. The database management system 8 includes a data management unit 80, a database 84 built in an internal or external storage device, and an external file 85.
Comprises a database control unit 81, a storage processing unit 82, and a restoration processing unit 83.

[0009] The database control section 81 receives a data control command issued through input means (not shown), and searches, inserts, and updates an object in the database 84. Further, the database control unit 81 is configured to save and restore the object in cooperation with the save processing unit 82 and the restoration processing unit 83.

The storage processing unit 82 includes a database control unit 8
In accordance with the object storage command from No. 1, all objects in a specified table in the database 84 are stored in the external file 85 via the database control unit 81. The restoration processing unit 83 restores the state of an object in the external file 85 saved by the saving processing unit 82 via the database control unit 81 to the database 84 based on an object restoration instruction from the database control unit 81. It is. In the database management system 8, the data management unit 80
Performs export and import of objects in the database 84 using the external file 85.

[0011]

In the above-described export and import in the ORDBMS, objects are saved and restored for each table. Therefore, even when only some of the objects in the table in the database are to be saved and restored, as shown in the example of FIG. 7, all the objects in the table are to be saved and restored. There was a problem that would be.

In addition, assuming a large-scale database, the execution of the export and import processes requires a great deal of time. Therefore, it is necessary to restrict access rights such as prohibition of access to the database during these processes. .

In a large-scale database, for example, even if it is desired to save only a set of objects related to a specific object in a file, in the export process, even objects that do not need to be saved are processed as objects to be saved. Therefore, there is a problem that the processing efficiency is greatly reduced.

Such a problem can be solved if the export and import processing can be performed for each specific object set, that is, for each specific data unit. Therefore, an object of the present invention is to provide a data management method that enables saving and restoring of data for each specific data unit. Another object of the present invention is to provide a data management device and a data management system suitable for implementing the above data management method. Another object of the present invention is to provide a recording medium for realizing the above data management method and apparatus on a general-purpose computer.

[0015]

A data management method according to the present invention for solving the above-mentioned problems is a data serial for searching related data from specific data in a searchable manner based on a data structure of a constructed database. And the data set to be stored is specified based on the data serialization information based on an external data storage request, and the specified data set to be stored is extracted from the database. And storing the data set stored in the first file in the database based on an external data restoration request, and storing the data set at the time of storage of the first file. Restoring, and managing the data group in the database based on the data serialization information. To reflect the association between over data mutually and executes it.

The data serialization information is, for example, information for arbitrarily and limitedly specifying a range of a data set to be stored to be spread from specific data as a starting point to related data. Also, the step of restoring the data set includes identifying information of each data at the time of saving the first file and identification information of each data when reflecting the data set saved in the first file in the database. Are restored in association with each other to maintain data consistency.

A data management apparatus according to the present invention for solving the above-mentioned other problems is an apparatus for saving and restoring data in a database, and stores a data group to be saved in accordance with an external data saving request in advance. Storage data specifying means for specifying based on the serialization information of the data group in the set database, data storage means for extracting the specified data group to be stored from the database and storing the data group in an external file, Data restoration means for reflecting a data group stored in the external file to the database in response to a data restoration request from outside, and restoring the data group at the time of storage of the external file. is there.

[0018] The stored data specifying means may, for example, convert one or a plurality of pieces of data related to a data storage request starting point as data to be stored based on serialization information of a data group in the database. It is configured to be specified sequentially.

The data storage means extracts, from the database, data to be specified and attribute information relating to the specified data, based on catalog information including definition information of a data group in the database. The external file is configured to be stored in a record format in the order specified sequentially from the data as a starting point. Alternatively, the identification information of the data to be saved sequentially specified from the data as the starting point is held, and the identification information corresponding to the completion of saving all the data to be saved to the external file can be deleted. The configuration is as follows.

For example, when the same data as the data group stored in the external file exists in the database, the data restoring means sequentially reads the data group in the external file and reads the read data in the database. And updating the data of the same identification information with the attribute information of the read data to restore the data. Alternatively, when newly generating the data group stored in the external file with respect to the database, sequentially reading the data group in the external file, and generating the same data as the read data in the database, It is configured to add new identification information to the generated data. Alternatively, the database includes an identification information correspondence table in which the new identification information given to the data generated in the database and the identification information at the time of storage of the data in an external file are associated with each other,
On the basis of the identification information correspondence table, the data to be restored is configured to maintain consistency between the storage time in the external file and the generation time in the database. Alternatively, the identification information correspondence table can be erased when the restoration of the data group stored in the external file to the database is completed.

[0021] Preferably, the data management device stores and restores a group of objects to which unique identification information is added in a predetermined object-relational database constructed by including a relational database including an object-oriented concept. Configure to do so. Alternatively, it is configured to save and restore an object group to which unique identification information is added in a database constructed as a predetermined object-oriented database.

According to another aspect of the present invention, there is provided a data management system comprising: a plurality of first devices for managing a data group in a database; and a plurality of first devices having external files for making data storage and restoration requests in the database. Are connected so that bidirectional communication is possible. In this configuration, the first device specifies a data storage request from the outside and serialization information of a data group in the database set in advance to specify a data group to be stored in accordance with the data storage request. Means for receiving from the second device, means for specifying a data group to be stored based on the received serialization information, extracting the data group from the database, and transmitting the extracted data group to the second device A data restoration request from the outside and a data group stored in the external file in the second
Means for receiving the data group from the device and reflecting the data group in the database to restore the data group at the time of saving the external file.

According to another data management system of the present invention, there are provided a plurality of first devices for managing a data group in a database, and a plurality of second devices each having an external file and performing a data saving and restoring request in the database. Are connected so that bidirectional communication is possible. In this configuration, the second device transmits a data storage request from the outside and serialization information of a data group in the database set in advance to specify a data group to be stored in accordance with the data storage request. Means for transmitting to the first device, means for receiving, from the first device, a data group respectively extracted from the specified database to be saved, and saving the data group in the external file in the own device. Means for transmitting a data restoration request from outside and a data group stored in the external file to the first device.

According to another aspect of the present invention, there is provided a recording medium storing a program readable and readable by a computer for storing and restoring data in a database.
At least the following processing is executed by the computer device. (1) a process of specifying a data group to be stored in response to a data storage request from the outside based on serialization information of the data group in the database set in advance;
(2) a process of extracting the specified data group to be stored from the database and storing the data group in an external file;
(3) A process of restoring the data group at the time of saving the external file by reflecting the data group stored in the external file on the database in response to an external data restoration request.

[0025]

Embodiments of the present invention will be described below in detail with reference to the drawings. (First Embodiment) FIG. 1 is a functional block diagram showing an embodiment of a data management device 1 to which the present invention is applied. This data management device 1 is an ORDB database 16 (hereinafter referred to as DB1) constructed in an internal or external storage device, with an RDB including an object-oriented concept.
6) and a database control unit 11 (hereinafter, referred to as a DB control unit 11) which is formed by a computer device reading and executing a predetermined program while having an external file 17 constructed for storing data in the DB 16. ), A storage processing unit 12, a restoration processing unit 13, a serialization information holding unit 14, and an OID management unit 15.

The above-mentioned program is usually stored in an internal or external storage device of a computer device, and is read and executed at any time. The computer device has any form capable of forming each of the functional blocks. It was recorded. For example, a portable recording medium such as a CD-ROM or FD that can be separated from the computer device,
Alternatively, the program may be recorded in a program server or the like connected to a network, read at the time of use, installed in the computer device or in an external storage device, and provided for execution at any time.

The DB control unit 11 receives a control command for the DB 16 from an operator or the like via an input interface (not shown), and searches, inserts, updates, etc. the DB 16 based on the control command. It is. The DB control unit 11 is configured to save and restore an object set in the DB 16 in cooperation with the storage processing unit 12 and the restoration processing unit 13.

To obtain a particular set of objects,
Serialization information is required. Therefore, the storage processing unit 12
Specifies the serialization information held by the serialization information holding unit 14 in response to the object set storage command from the DB control unit 11, and based on the serialization information, the OID management unit 15 Use DB
Object set extracted from the external file 1
7 and save it.

The restoration processing unit 13 converts the object set in the external file 17 stored by the storage processing unit 12 into the DB control unit 11 and the OID management unit 15 in response to the object set restoration command from the DB control unit 11. The object status is restored by importing it into the DB 16 and reflecting the status.

The serialization information holding section 14 holds information including serialization rules and the like in the object set in the DB 16. The serialization information regarding the object set will be described later.

The OID management unit 15 manages OID, which is identification information on an object group in the DB 16. Specifically, in the case of export, O
The ID information is stored based on the OID management table, and in the case of import, the object is restored based on the OID correspondence table in which the old OID and the new OID are associated with each other. The OID management unit 15 will be described later.

The external file 17 is constructed for saving, ie, exporting, specific data including objects in the DB 16, and is configured using, for example, storage means such as a memory or a disk device.

Next, serialization of an object set in the data management device 1 of the present embodiment will be described with reference to the drawings. The serialization information holding unit 14 in the present embodiment holds the serialization information for serializing the object sets, that is, the ordering of related object sets for storing the set of objects in the DB 16 in the external file 17. Things.

Based on the serialization information, the storage processing unit 12 searches for a related object from a specific object serving as a starting point of serialization, serializes the object, and stores an object set in the external file 1.
7 is stored. Hereinafter, such an object serving as a starting point is referred to as a “starting point object”, and a restoration process for the DB 16 from a serialized object set stored in the external file 17 is referred to as “deserialization”. Described and explained.

In general, in order to perform serialization, it is necessary to provide a rule for searching for a related object from a given starting object. In this embodiment, first, the range of a class to be stored is specified. To identify
The serialization information holding unit 14 holds a serialization rule table set in advance. The rule table specifies, in the storage processing unit 12, after storing an object of a specific class, which class object related to the object should be stored thereafter. That is,
When a specific object is serialized, this is information for specifying which class of object should be serialized next from the rule table.

For example, in a data model having a class structure as shown in FIG. 2, a solid line between each of the classes A to E in FIG.
The numbers and “*” are multiplicity indicating the degree of association between the classes. For example, there is a one-to-one relationship between class B and class C, and class A and class B
This indicates that there is a one-to-many relationship with. FIG.
Is a construction example in which the data model is realized in the DB 16. In the figure, the association between the classes is realized by “reference” indicated by a wavy arrow between the tables A_TBL to E_TBL respectively corresponding to the classes.
Note that the direction of reference between tables is not uniquely determined from the class structure. That is, when a data model having a class structure as shown in FIG. 2 is given, a table that can realize the data model includes:
There will be multiple design methods.

Here, a serialization rule for realizing the examples of FIGS. 2 and 3 is determined, for example, as follows. 1. If a class A object is saved, the related class B object is also saved. 2. If a class B object is saved, the related class C object is also saved. 3. If a class B object is saved, the related class D object is also saved. 4. If a class C object is saved, the associated class E object is also saved. 5. If a class D object is saved, the associated class E object is also saved.

According to the above serialization rule, for example, when an object in class B is set as a starting object and an object of another class related thereto is serialized, first an object of class C, then an object of class D , And finally an object of class E. It should be noted that there is no rule that "if an object of class B is saved, an associated object of class A is saved". Thus, when the class A object is saved, the class B object is saved, but when the class B object is saved, the class A object is not saved. In other words, when determining the serialization rule, it is always determined that the effect is to be applied to the unidirectional object, and the serialization rule table in the serialization information storage unit 14 is generated and stored in advance. Good.

Since it is not clear from the physical structure in the DB 16 what rule is used for serialization, an abstraction as shown in FIG. 2 is extracted in order to extract a semantically related object set. A user or the like may appropriately determine a serialization rule based on a data model having a high class structure. From this, the serialization rule reflects the association between the class and the object.

An implementation that automatically converts a serialization rule determined by a user or the like based on an abstract data model into, for example, a serialization rule table including correspondence between class names and table names in the DB 16. What is necessary is just to realize using an environmental tool.

The serialization rule table is a set of binary relations having class names as elements, or a directed graph having class names at vertices.
If the table names in are associated one-to-one,
The serialization rule table corresponding to the serialization rule can be represented as <Serialization Rule Table> shown below. This table is a correspondence table between a table name including a starting point object, ie, a “starting point table” and a table to which serialization should spread from the table, ie, a “spreading table”.

<Serialization Rule Table> (Starting Point Table) (Spread Table) A_TBL B_TBL B_TBL C_TBL B_TBL D_TBL C_TBL E_TBL D_TBLE E_TBL

Next, catalog information in the DB 16 will be described. The catalog information is information for managing the DB 16, and includes definition information about a table, specifically, a table name and a set of attribute definitions included in the table. For example, <Table 1> shown below is an example of definition information regarding tables B_TBL and C_TBL. here,
“B_TBL” and “C_TBL” are table names,
Below the table name is the attribute definition. An attribute definition is defined by an attribute name and its type. For example, in the case of table B_TBL, "n
"ame", "age" and "birthday" are shown in Table B_
Attribute in the TBL, while "addres
s "," ref_B_TBL "," ref_E_TB "
“L” is an attribute in the table C_TBL. This attribute must define the types of values that can be taken. For example,
“Name” is a character example of 20 characters, and “age” is a positive integer. Here, Table C_
TBL attributes "ref_B_TBL" and "ref_E_
The type of “TBL” is the OI of the object in the specified table.
Take D. For example, "ref_B_TBL" corresponds to table B_
The OID of the object stored in the TBL is entered. By referring to this catalog information, it is possible to know which table object an object in a certain table refers to, that is, the direction of reference between tables.

[0044]

FIG. 4 is a block diagram showing an example of a storage format of the serialized object set in the external file 17. External file 17 in this embodiment
Then, the storage processing unit 12 is configured in a column format including, for example, the OID of the object, the table name to which the object belongs, and the attribute value of the object for each object. In this case, the table name only needs to uniquely identify the table. For example, a unique identifier may be assigned to the table name, and the table name may be substituted by the identifier.

Next, restoration of an object set in the restoration processing unit 13, that is, deserialization will be described. The restoration processing unit 13 according to the present embodiment
An object set in an ordered state, which is serialized and stored in the database, is imported into the DB 16, and the object set is deserialized and restored. In this case, when an object existing in the DB 16 is updated (overwritten) by the serialized object set in the external file 17, the same OID as that of all the serialized object sets is used.
16 has already been assigned to the object existing on
It is also assumed that the reference object of the serialized object set exists on the DB 16.

On the other hand, when a serialized object set is newly generated on the DB 16, all serialized object sets are generated on the DB 16 together with a new OID. At this time, it is necessary to prevent OID inconsistency due to object reference.

The OID management unit 15 of the present embodiment stores the OID (old OID) of the serialized object set in the external file 17 and the OID in the DB 16.
(New OID) is provided with an OID correspondence table associated with each object, and OID management is performed to maintain the consistency of the objects related to the restoration processing.

First, when an object set in the external file 17 is newly generated on the DB 16, all the OIDs are reassigned in the DB 16 when the restoration processing unit 13 restores the object set. . At this time, the OID management unit 15 generates an OID correspondence table that associates the old OID at the time when the object set is serialized with the newly reassigned new OID. Specifically, this OID correspondence table shows the old OI in all object sets at the time of serialization.
D and a new O corresponding to the restoration process of the object set
IDs are associated with each other. This allows
The object set newly generated in the DB 16 can be restored based on the reference relationship between the objects at the time of serialization.

Next, a specific processing procedure in the data management device 1 of the present embodiment will be described with reference to the drawings. In the following, a case will be described in which a data model having the class structure shown in FIG. 2 is constructed and realized in the DB 16 as shown in FIG. Also,
The tables A_TBL to E_TBL in FIG. 3 store the objects of the classes A to E in FIG. 2, respectively. The case where the object is stored will be described. The above-mentioned <serialization rule table> is used for the serialization rule.

In the present embodiment, an OID management table is used. This OID management table is used by the OID management unit 15 to prevent the same object from being stored more than once when serializing an object set. To
This is a temporary work table for storing the OIDs of the stored objects. For example, the OID management table 15 is configured to be deleted after the object storage processing is completed in the OID management unit 15.

FIG. 5 is a diagram showing a procedure for storing an object set in the data management apparatus 1. First, the save processing unit 12 extracts and acquires a starting object specified by the user from the DB 16 in cooperation with the DB control unit 11 in accordance with an object set saving instruction from the DB control unit 11 (step S101).

Next, the saving processing unit 12 exports the acquired starting object to the external file 17 and saves it (step S102). In this case, the OID of the stored origin object is
It is held in the ID management table. The storage processing unit 12 specifies a table to be serialized thereafter based on the serialization rule table in the serialization information holding unit 14 (step S10).
3) If there is no table to be serialized (step S104: No), the process ends.

On the other hand, if the table to be serialized is specified (step S104: Yes), the DB 16 specifies the reference direction from the origin object to the table to be serialized (step S105). For example,
The origin object b3 in the table B_TBL is the table C_
The storage processing unit 12 searches for objects in the tables C_TBL and D_TBL related to the origin object b3 in cooperation with the DB control unit 11 because the objects are related to the objects in the TBL and D_TBL. Specifically, in order to specify the object in the table C_TBL related to the origin object b3, the DB control unit 11
Specifies the reference direction of the object from the table C_TBL to the table B_TBL by referring to the catalog information in the DB 16.

Next, the DB control unit 11 searches for an object related to the origin object based on the specified reference direction (step S106). Specifically, an object C3 related to the origin object b3 is acquired from a reference to the table B_TBL, which is one attribute value in the table C_TBL. In the case of the table D_TBL related to the origin object b3, the related objects d1, d2, and d3 are obtained in the same manner.

Further, the DB control unit 11 determines whether or not there is any stored object in the object group related to the acquired starting object based on the OID management table in the OID management unit 15 (step S10).
7). If the object is a stored object (Step S107: Yes), the process returns to Step S103, and the DB control unit 11 specifies a table to be serialized next.

On the other hand, if the object is not a stored object (step S107: No), the process returns to step S102, where the storage processing unit 12 exports and saves the object to the external file 17 and stores the object in the OI.
In the D management unit 15, the OID of the object is
Insert into the management table. By the above procedure, external file 1
7, an object set composed of serialized objects is stored. When the saving process for all the related objects is completed, the OID management unit 15 is configured to appropriately delete the OID management table.

FIGS. 6 and 7 are flowcharts showing a process of restoring an object set in the data management apparatus 1. FIG. FIG. 6 is a processing procedure diagram in the case where an object existing on the DB 16 is updated by a serialized object set in the external file 17. The restoration processing unit 13
In response to an object set restoration command from the control unit 11,
One object is read from the head of the external file in which the serialized object set is stored (step S201). While the external file 17 is not the end of the file (step S202: No), the restoration processing unit 13
Cooperates with the DB control unit 11 and the OID management unit 15,
Same as object read in step S201
The object in the DB 16 to which the ID is assigned is updated by overwriting the attribute value of the read object (step S203), and the process returns to step S201 to repeat the process. On the other hand, external file 1
7 is the end of the file (step S202: Ye
s), end the process.

FIG. 7 is a processing procedure diagram when a serialized object set in the external file 17 is newly generated on the DB 16. The restoration processing unit 13
In response to the object set restoration command from the B control unit 11, one object is read from the head of the external file in which the serialized object set is stored (step S301).

While the external file 17 is not the end of the file (step S302: No), the restoration processing unit 13
The B control unit 11 determines whether the read object has an attribute that refers to another object or not.
The determination is made based on the catalog information in S16. If the determination result indicates that there is an attribute that refers to another object (step S303: Yes), the DB control unit 11
Detects whether the OID related to the attribute value exists in the OID correspondence table in the OID management unit 15, and if the OID exists in the OID correspondence table (step S304:
Yes), the OID management unit 15 converts the OID into a new OID in the OID correspondence table and changes the attribute value to a new OID.
ID (step S305).

If the new OID does not exist in the OID correspondence table (step S304: No), the DB control unit 11 generates an object to be referred to corresponding to the new OID in the DB 16, and generates the object. A new OID is assigned to the created object. Also, the OID management unit 15
Then, for the generated object, the new OID is associated with the old OID at the time of saving the external file in the OID correspondence table (step S306).

Next, the DB control unit 11 sets the new OID of the object in the DB 16 as one attribute value of the corresponding object (step S307). After completion of step S307, or when the determination result in step S303 indicates that there is no attribute that refers to another object (step S30)
3: No), it is determined whether or not the new OID for the OID exists in the OID correspondence table in the OID management unit 15. If the OID exists in the OID correspondence table (step S308: Yes), the DB control unit 11 sets each attribute value for the object corresponding to the new OID in the DB 16 (step S309).

On the other hand, if the OID does not exist in the OID correspondence table (step S308: No), the DB control unit 1
1 generates an object of the new OID in the DB 16 and sets each attribute value, and the OID management unit 15 stores an external file 1
7, the old OID at the time of saving and the new OID generated in the DB 16 are associated in the OID correspondence table (step S310). The OID correspondence table is, for example, as shown in Table 2 below.

<Table 2> (Old OID) (New OID) Old OID of b3 New OID of b3 New OID of old OID of e1 New OID of c3 New OID of old OID c3

After the completion of step S309 or 310, the process returns to step S301 to repeat the processing. From the above Table 2, the OID correspondence table for the objects d1, d2, and d3 related to the origin object b3 is as shown in Table 3 below.

<Table 3> (Old OID) (New OID) Old OID of b3 New OID of b3 New OID of e1 New OID of c1 Old OID of c3 New OID of c3 Old OID of d1 Old OID of d1 New OID of OID d2 New OID of old OID d3

If the external file 17 is the end of the file (step S302: Yes), the OID management unit 1
In step 5, the corresponding OID correspondence table is deleted (step S5).
311) End the process.

In the present embodiment, the case where the present invention is applied to saving and restoring of an object set in an ORDBMS has been described. However, the present invention is not limited to the above-described embodiment. Applicable.

Further, the present invention is not limited to the construction of the database as in the above embodiment as long as the data serialization information can be generated so that the data storage range can be specified for the data group in the database. It can be applied to

As described above, the data management device 1 of the present embodiment
Then, based on the serialization rules for the objects in the database, compared to saving and restoring objects in units of tables as in the conventional method, all objects in the table are not saved and restored,
It is possible to arbitrarily set only a range of some objects in a table in a database as a target of saving and restoring.

Also, since the serialization rule is based on the reference relation of objects in a series of related object sets specified in the database, the object set is stored in an external file or the like while maintaining the reference relation. It is possible to save and restore the object set database from the external file.

Further, an object set stored in an external file or the like can be additionally inserted into another database having the same table structure with a new OID serving as an object identifier.

As described above, unlike the export processing of the conventional method, only the object set related to a specific object can be saved in the external file without making the objects that do not need to be saved the object to be saved. Export and import processing becomes possible, and practical use under a large-scale database becomes possible.

Further, by the quick export and import processing, for example, the time limit for the access right during the processing is reduced, and the convenience and processing efficiency of the user are greatly improved.

(Second Embodiment) The present invention relates to a plurality of computer devices connected to each other via a public network such as the Internet as a communication line, for example, a plurality of database servers which are data storage devices. , And a plurality of data management servers, which are data management devices, can be implemented as a data management system.

The data management server in this case is, for example,
Data management is performed for a plurality of large-scale databases in a network environment. As an example of the configuration, the same external file as the external file 17 is constructed inside a computer device or on an external storage device, and a public network is constructed. A communication control unit for performing data communication with the database server via the DB, a DB control unit 11 similar to the data management device 1, and a storage processing unit 1.
2. It comprises a restoration processing unit 13 and a serialization information holding unit 14.

Similarly, the database server is provided with a communication control unit for performing data communication with the data management server, and a database similar to the DB 16 in the data management device 1 is constructed in its own device. 1
1 and the OID management unit 15.

This data management system is different from the data management device 1 in that the data management system includes a communication control unit for performing communication control. A specific data group to be transmitted to the database server on the public network via the public network and to be stored from the database server is received via the communication control unit. In the serialization information holding unit 14 in this data management server,
What is necessary is just to generate and hold a plurality of serialized information for each database server in advance, and the serialized information may be configured to be transmitted together with a data storage request to the corresponding database server. Further, the serialized data group stored in the external file 17 may be configured to be transmitted to the database server together with the data restoration request.

Further, in the database server, a data storage request, a restoration request, and serialization information from the data management server are received via the communication control unit and input to the DB control unit 11. Transmits a data group extracted based on the received serialization information to the data management server via the communication control unit. Also, in the case of a data restoration request, the data group to be restored from the data management server is received via the communication control unit and input to the DB control unit 11, so that the substitution can be performed. The same effect as that of the data management device 1 can be obtained.

Thus, in the data management system of the present embodiment, it is possible to construct a system that can optimally cope with a large-scale database that is remotely deployed, shorten the processing time for data management, and reduce the system time. Overall processing costs are greatly reduced.

[0081]

As is apparent from the above description, according to the present invention, there is a specific effect that data can be saved and restored for each specific data unit. This allows
For example, in a relational database constructed based on the object-oriented concept, there is an effect that an object set can be stored in an external file or the like and an object set can be restored from the external file or the like.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a data management device according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram showing an association between classes.

FIG. 3 is a conceptual diagram showing a reference relationship between a table and objects.

FIG. 4 is a block diagram showing a storage format of a serialized object set in the embodiment.

FIG. 5 is a processing procedure diagram illustrating a saving process in the embodiment.

FIG. 6 is a processing procedure diagram illustrating restoration processing in the embodiment.

FIG. 7 is a processing procedure diagram showing a restoration process in the embodiment.

FIG. 8 is a functional block diagram of a conventional database management system.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 data management device 8 database management system 11 database (DB) control unit 12 storage processing unit 13 restoration processing unit 14 serialization information holding unit 15 OID management unit 16 database (DB) 17 external file 80 data management unit 81 database control unit 82 Storage processing unit 83 Restoration processing unit 84 Database 85 External file

Claims (16)

[Claims] 1. Data serialization information for exploring and identifying related data from specific data based on a data structure of a constructed database, and based on a data storage request from outside, Specifying a data set to be stored based on the data serialization information, extracting the specified data set to be stored from the database, and storing the data set in a first file; Reflecting the data set stored in the first file in the database based on the data file, and restoring the data set at the time of saving the first file. Characterized in that it is executed by reflecting the relationship between the data based on the data serialization information, Data management method. 2. The data serialization information is information for arbitrarily and limitedly specifying a range of a data set to be stored to be spread from specific data serving as a starting point to related data. The data management method according to claim 1, wherein: 3. The step of restoring the data set includes identifying the identification information of each data at the time of storage of the first file and the data set when the data set stored in the first file is reflected in the database. 2. The data management method according to claim 1, wherein the identification information is associated with each other and restored to maintain data consistency. 4. An apparatus for saving and restoring data in a database, wherein a data group to be saved in response to a data saving request from the outside is converted into a predetermined serialization information of the data group in the database. Storage data specifying means for specifying based on the data, data storage means for extracting the specified data group to be stored from the database and storing the extracted data group in an external file, and the external file in response to a data restoration request from the outside And a data restoring unit for restoring the data group at the time of saving the external file by reflecting the data group stored in the external file on the database. 5. The storage data specifying means, based on serialization information of a data group in the database,
Related 1 from the data that is the starting point of the data storage request
5. The data management device according to claim 4, wherein the data management device is configured to sequentially specify a plurality of data as data to be stored. 6. The data storage means extracts data to be stored and attribute information on the specified data from the database based on catalog information including definition information of a data group in the database, and The data management device according to claim 4, wherein the data management device is configured to store in the external file in a record format having an order sequentially specified from data as a starting point. 7. The data storage means holds identification information of data to be stored sequentially specified from the data as a starting point, and triggers completion of storage of all data to be stored in the external file. The data management device according to claim 4, wherein the identification information corresponding to the data is configured to be erasable. 8. When the same data as the data group stored in the external file exists in the database, the data restoring unit sequentially reads the data group in the external file, and reads the read data in the database. Data of the same identification information as
The data management device according to claim 4, wherein the data management device is configured to update the read data with the attribute information and to restore the data. 9. The data restoring unit, when newly generating a data group stored in the external file with respect to the database, sequentially reads the data group in the external file, and stores the read data and 9. The data management device according to claim 8, wherein the same data is generated in the database, and new identification information is added to the generated data. 10. The data restoration means according to claim 1, wherein said new identification information given to the data generated in said database is associated with identification information of said data at the time of storage in an external file. Comprising a table, based on the identification information correspondence table, configured to maintain consistency between the storage time of the external file and the generation time of the database in the external file. The data management device according to claim 8, 11. The data restoring means is configured to be able to delete the identification information correspondence table upon completion of restoration of a data group stored in the external file to the database. Item 11. The data management device according to Item 10. 12. A data management apparatus according to claim 4, wherein the identification information unique to a predetermined object-relational database constructed by including an object-oriented concept in a relational database is provided. A data management device configured to save and restore an assigned object group. 13. A data management apparatus according to claim 4, wherein said data management apparatus stores and restores a group of objects to which unique identification information is assigned in a database constructed as a predetermined object-oriented database. A data management device, characterized in that the data management device is configured to perform the following. 14. A plurality of first devices for managing a data group in a database, and a plurality of second devices having an external file and performing data saving and restoring requests in the database are connected so as to be capable of two-way communication. The first device transmits a data storage request from outside and a serialization information of a data group in the database set in advance to specify a data group to be stored according to the data storage request. Means for receiving from the second device, a data group to be stored is specified based on the received serialization information, extracted from the database, and the extracted data group is transmitted to the second device. Means for receiving, from the second device, a data restoration request from the outside and a data group stored in the external file,
Means for reflecting the data group on the database and restoring the data group at the time of saving the external file. 15. A plurality of first devices for managing a data group in a database, and a plurality of second devices having external files and requesting data storage and restoration in the database are connected so as to be capable of bidirectional communication. The second device includes a data storage request from the outside and serialization information of a data group in the database set in advance for specifying a data group to be stored in accordance with the data storage request. Means for transmitting to the first device, means for receiving, from the first device, a data group respectively extracted from the specified database to be saved, and saving the data group in the external file in the own device. Means for transmitting a data restoration request from outside and a data group stored in the external file to the first device. Data management system consisting of Te. 16. A recording medium for storing and restoring data in a database, in which a program readable by a computer device is recorded, wherein the program is a storage medium in accordance with at least an external data storage request. A process of specifying a data group to be stored based on the serialization information of the data group in the database set in advance, a process of extracting the specified data group to be stored from the database and storing it in an external file, Reflecting the data group saved in the external file in response to the data restoration request from the external file to the database and restoring the data group at the time of saving the external file. A recording medium, characterized in that: