Learnaboutgmp Community

Performance Qualification

We are implementing and validating a CDMS. Have a clarification on the Performance Qualification to be performed to execute a ‘good’ CSV. We have been provided a “full functionality” test script by the CDMS vendor that checks every possible functionality of the CDMS. We plan to run this full functionality test script on our TEST INSTANCE of CDMS. Do we need to execute the full functionality on PRODUCTION INSTANCE TOO??

Also for every new protocol/study we plan to pick a sub set of the full functionality test script and execute as PQ. Does this study specific PQ need to be executed on TEST AND PRODUCTION INSTANCE? If this is executed on PRODUCTION we would need to provide a dummy name for the same to ensure it does not conflict with the actual study.
Can some one help??

Have you got a test to show that the test system is identical to your
production system, if so there is no need to retest all functionality in production. What I would do is perform some degree of PQ testing to ensure the system is working correctly and consistently once live.

[quote=pradeeptonpe]
Also for every new protocol/study we plan to pick a sub set of the full functionality test script and execute as PQ. Does this study specific PQ need to be executed on TEST AND PRODUCTION INSTANCE? If this is executed on PRODUCTION we would need to provide a dummy name for the same to ensure it does not conflict with the actual study.
Can some one help??[/quote]

I’m not sure if I fully understand this part of the question, but I would pick a section of the PQ that tests a fully process run if possible.[/quote]

Please explain further

Your response to the first part of the question addresses the second part too. thanx!
Add on clarification: How can we test/prove that two instance (say test and prod) are identical??

Your response to the first part of the question addresses the second part too. thanx!
Add on clarification:

[quote=pradeeptonpe]
How can we test/prove that two instance (say test and prod) are identical??[/quote]

I would show that the source code or executable file is the same version as what is on your test system.

I asume that you have this executable stored in a secure location, and when updates are made to the source code this is reflected in the version.

It’s all about version control of your software and how it is maintained.

Perhaps reference your version control procedure if you have one.

Regards