Recommendations for migrating applications from a test to a production environment

When you migrate applications that are enabled with pureQuery client optimization from a test environment to a production environment, follow the recommendations for best results.
It is recommended to run capture and test in an environment that is as similar as possible as the final target environment. However, in many cases, it might not be possible for the test and production environments to match. Examples of the differences between a test and a production environment include:
  • The test systems might use a DB2® for Linux®, UNIX®, or Windows® database, but the production environment uses a DB2 for z/OS® database.
  • The test systems might use a different version of the database than the production systems
Before you migrate an application, you must ensure that schema information matches in the two environments. Both environments can use schemas in one of the following ways:
  • Both systems use the same schema definitions.
  • Both systems use unqualified table names.
On the production system when you switch to executing SQL statements statically from executing them dynamically, you must test the results. Examples of items to test include:
  • Ensure that the bind operation on the target database is successful.
  • Ensure the static execution of SQL statements return the same results as the dynamic execution of the statements.
  • Ensure that the application performs in the same manner when executing the statements statically as when the statements are executed dynamically.

You might need to reconfigure the pureQueryXML file with production-specific attributes like the package names and collection. After you run the Configure utility on the pureQueryXML file with production-specific options, the StaticBinder can be used to create the final packages on the production system.

z/OS encoding considerations

The pureQuery Runtime utilities operate on a UTF-8 encoded files. When you capture SQL data, the pureQueryXML file that contains the SQL data is created in the UTF-8 encoding. A z/OS UNIX System Services command, such as vi, that is used to open a UTF-8 encoded file might not work.

You might need to run the iconv command to convert the encoding for this file. For example, to convert the file capture.xml to IBM-1047 encoding, use the following command:
iconv -f UTF-8 -t IBM-1047 capture.xml > capture_1047.xml

After the conversion, you can use vi to open and view the contents of capture_1047.xml. However, you cannot use the converted file with pureQueryRuntime. You must use a file with UTF-8 encoding.

The pureQuery Runtime utilities operate on a UTF-8 encoded files, and so the original file capture.xml must be provided to the pureQuery Runtime utilities.

If you transfer a file by using FTP from a z/OS system to a workstation for viewing and processing, you must transfer the file in binary mode. Also, ensure that you are using an editor that can edit UTF-8 encoded files.


Feedback