Using Oracle’s Data Conversion Tool—DCT

Oracle offers a product called the DCT. More than being an super-advanced piece of software, it is encapsulated knowledge from Oracle’s experiences with migration. What you are really purchasing with DCT is automated experience in the migration process.

What the Conversion Tool Is and What It Is Not

Oracle’s Data Conversion Tool, the DCT, is not a magical box that will allow you to move any application from any environment to Oracle. It is, instead, a methodology centered around a series of conversion tools. The major tools within the DCT are the following:

•  Schema Reference

•  Flat File Layout

•  DDL Generator

•  SQL Loader Generator

•  Conversion Mapper

•  Object Deployer

•  Conversion Generator

•  Impact Analyzer

The DCT is run from a new, open-system environment such as UNIX or Windows NT, and has an easy-to-use graphical front-end. From the database the DCT uses, you can create a central repository for the schemas used in both the old and new systems.

Data mapping is defined between your older system and the Oracle system you are migrating to when using this tool. The DCT will also generate simple SQL*Loader scripts and COBOL data-mapping information as a template for the conversion.

The methodology of the DCT is to create a staging area for data. First, data is moved from the mainframe to this staging area by using extract files created by the DCT. After this is done, data is moved into an Oracle database that supports the staging area. PL/SQL then translates data from the staging area into the new schema that exists in the DCT repository (see Figure 7.13).

Figure 7.13.  The DCT uses a migration methodology to create the key elements needed in a migration effort, and a way to keep track of these elements.

Even with this tool, much of the coding still needs to be performed manually by a programming analyst. That is why the DCT creates many shells of PL/SQL procedures, all within a conversion PL/SQL package. These statements need to be modified and filled in with logic because they involve specific mapping rules unique to your migration effort.

What the DCT gives you above all else is a way to manage and keep track of all the data schema details in a large migration effort. It also provides standard templates and a step-by-step methodology for the migration of data from a non-Oracle database to an Oracle database.

A Step-by-Step Tour of a Data Migration Using the DCT

Follow the step-by-step methodology of the DCT to better understand its components. Along with learning about what automation the DCT offers, you will understand the methodology that Oracle is selling with the tool. This methodology is a very clear and straightforward way to migrate a database to Oracle.

Step One: Identify and Load Schema Data

In this step, DCT basically is used by data analysts to plan how data is going to be moved into the new Oracle database from the existing system(s). Obviously, this process will, to a great degree, drive how the new Oracle database is designed.

When using DCT, you can opt to purchase SQL*Net and/or a Transparent Gateway. You can directly link the new database to the older systems with these expensive tools, and DCT can perform an almost automatic mapping. When you choose this path, you cause the old system’s data to appear to Oracle as an Oracle database, so many of the steps to download and massage data can be avoided. What you are really doing in this case is moving data from a virtual Oracle database to an actual Oracle database (see Fig- ure 7.14).

Figure 7.14.  If you chose to use DCT in conjunction with SQL*Net and/or a Transparent Gateway, DCT can automatically determine the older system’s schema.

Step Two: Identify and Load Legacy Extract Files

If you are not so fortunate as to have transparent gateways between your old system and the new Oracle RDBMS, you need to extract data from the old system into a flat file by using either the other database vendor tools or a language such as COBOL.

DCT will automatically use a COBOL file layout to generate schema elements. Other languages can also be used and added to the DCT to automate the data definition process. A user can also manually enter flat-file layouts using the Flat File Layout (FFL) screen.