Table Copying

Fast & reliable test data provision
Corporations maintain large amounts of data in relational databases such as Oracle, SQL Server, Db2/LUW, etc., and it is normal to find several instances for production, test and development. Smoothly running daily operations can only be guaranteed through intensive testing over the course of the development cycle and final regression. Quality assurance requires test data like the data in production. The test data needs to be refreshed frequently and regularly. In order to increase operational safety, separate DBMS instances for production, test and development need to be introduced. Refreshing test data means copying data from production instances over to test and development instances.

The time window for consistent migration of production data decreases as the need for 24×7 operations increases. Limited resources, as well as shrinking batch windows, impede the transfer of test data. In instances where the test data must be completely refreshed, the task becomes relatively simple. In this case, it essentially comes down to efficiency and copy speed.

However, the task is usually more challenging. On the one hand, test data can already be present which will never be overwritten. For example, manually entered test cases that do not occur within the current production data yet because they are used to test newly created functions. On the other hand, data structures could differ in test and development from the current structures in production. These new structures have to be maintained and should, wherever possible, be filled with current production data. Often differing object identifications between test and production must also be taken into account. Therefore, flexible replacement and refresh options are required.

Both the naming pattern, as well as the exact comparison of the source and target objects must be considered. This explains why the manual provision of test data is quite complicated, time consuming and error prone, and, as a result, expensive. Both the complex, error-prone checks and preparations are automated by the copy tool XDM Table Copying. Plus, the execution itself is accelerated and transferred to a robust, restart capable process.


XDM Table Copying quickly and efficiently copies, updates and replicates databases, tables, views, etc. It can copy within an instance as well as to other instances, even to instances of other DBMS types. A copy job integrates the physical data movement and the processing of the data structure (DDL). It will copy individual tables or numerous databases and creates the objects in the target instance if they do not already exist. XDM Table Copying always selects the fastest and most efficient way to move the data. This way copy jobs are completed in record time. DBAs are no longer forced to wait for a time window over the weekend, they are now free to copy during regular business hours. Manual checks and preparation work are things of the past. No expert knowledge is required for the execution of established copy jobs.

The dream of test data provision on demand has finally come true for QA. XDM Table Copying significantly cuts down runtime & costs. It automates all copy procedures and reduces manual efforts by more than 90%. The workstation is intuitive to use, no specific knowledge of the particular DBMS is required. The user must only specify selection, exclude and renaming rules for a schema, a database or a set of tables and define the refresh options (fully replace, add non existing, etc). After that, the copy process runs fully automated and can be processed either step by step, or entirely at once. A XDM Table Copying task executes either under the control of the product or an in-house scheduler. Once everything has been defined and the copy task has begun, the process doesn‘t require any additional manual work or attendance. XDM Table Copying automatically generates the DDL for the selected objects/databases, respecting the predefined naming conventions of the target environment. The DDL is executed
in the target system and generates all the required instructions for the copy program. Structural differences are identified and automatically considered.

Why is XDM Table Copying so fast?
The copy program works in multi-thread mode, copying tablespaces in parallel. Nothing is required of the user. The scheduler “sees” and starts only one single process.

XDM Table Copying in a nutshell
XDM Table Copying integrates and automates the entire workflow of the data migration between relational DBMS. It is also possible to extract only the DDL of the source
objects and to establish “empty” environments in the target
system. A platform-independent script language is available to create and process copy tasks, or to conduct
considerable changes to existing tasks. XDM Table
Copying tasks are autonomous and can also be executed unattended by a standard scheduler.

XDM Table Copying supports the popular platforms and relational database management systems (RDBMS) including:

  • Db2 for z/OS, for Linux, Unix, Windows and AIX
  • Oracle for Windows, Unix and AIX
  • Microsoft SQL Server