Large volumes of data created by corporate activities have been accumulating in databases on mainframe systems. Recently, many companies have been distributing data from their mainframe systems to workstations (WSs) and personal computers (PCs) by employing a database management system (DBMS), such as HiRDB, that runs on workstations or personal computers. As this type of system operation becomes increasingly popular, questions of how to integrate and utilize distributed databases become important issues.

The facilities for importing data from a distributed database into another database are called replication facilities. Replication facilities enable you to import data between databases in different systems, thereby providing support for data management in a distributed system environment.

There are two types of replication facilities:

Data linkage facility

You use the data linkage facility to extract data from a database that has been updated and to automatically import the database update information into another system's database. This process of automatically importing data into another system's database is called data linkage. A database system equipped with the data linkage facility is called a data linkage system.

The data linkage facility is useful when you use the most recent data of one database in another system's database, and when you automatically back up databases.

Database extraction/import service facility

You use the database extraction/import service facility to extract data from a database in the batch mode and to import data into a database in another system. The database extraction/import service facility is useful when you extract and store a large amount of data in batch mode, and when you create and re-create databases in other systems.

The following table lists the products required for using the replication facilities with HiRDB and mainframe databases.

Table 1-1 Products required for using the replication facilities

Replication facility	HiRDB data linkage product	Data linkage product for mainframe database
Data linkage facility	HiRDB Datareplicator	XDM/DS#2
Database extraction/import service facility	HiRDB Dataextractor#1	XDM/XT#3

#1

For details about HiRDB Dataextractor, see the HiRDB Dataextractor Version 8 Description, User's Guide and Operator's Guide.

#2

For details about XDM/DS, see the VOS3 XDM/DS manual.

#3

For details about XDM/XT, see the VOS3 XDM/XT manual.

One way of linking data between two databases is to update one database whenever the other database is updated. The drawback to this method is that it requires a very heavy system workload for synchronization, error recovery, and so on. In contrast, data linkage using Datareplicator reduces the system workload because it extracts update information from one database, and then uses that data for asynchronous updating of a different database in another system.

Data linkage using Datareplicator is effective when performance and reduction of overall system workload are more important than precise synchronization of the distributed databases.

Datareplicator's data linkage facility also provides the following benefits:

• There is no need to modify applications, because Datareplicator extracts update information automatically.
• The data linkage facility transfers only the updated data, thereby minimizing the amount of transfer data and the key system's workload.
• You can create customized tables tailored to the needs of the various data-warehouse users, because the data linkage facility allows you to, among other things, select the tables and columns that will be subject to update processing and accumulate update information in chronological order.

The following table defines the terminology for data linkage systems used in this manual.

Table 1-2 Terminology for data linkage systems

Terminology for data linkage system	Definition
Source system	The system from which the updated data to be imported is extracted (linkage source)
Source Datareplicator	Datareplicator running in the source system
Source database #	Database in the source system
Target system	The system into which the extracted update data is imported (linkage target)
Target	Datareplicator running in the target system
Target database #	Database in the target source system

#

To identify a source database or target database by the system in which it is running, a HiRDB database is called a source HiRDB database or a target HiRDB database, as appropriate, while a database at a mainframe system is called a mainframe database.