I'm working for a company looking to introduce QC and QTP into there System Test and QA testing environments.
These environments operate independently. System Test will be performed by IT Development prior to QA. This is an attempt by the company to improve the quality of code delivered to QA. QA will be performed by dedicated testers. At present all testing is done manually and QTP will be introduced to implement automation.
How should I setup QC
1) One QC with 2 domains: 1 for ST and 1 for QA with projects hanging off these.
2) One QC with many domains for functional areas. Hanging of these would be projects some that would be QA anmd some ST.
3) Multiple QC's, One for QA and 1 for ST. Domains for each functional areas and various projects hanging off from these.
Option 3 appears to be the most logival option.
1) Is it possible to install multiple QC instances, what are the pitfalls and advantages. Are there any licencing issues.
2) Can object repositories be exported/imported between QC instances.
Many thanks for any help and advice I receive.
First of all QC's architecture will allow you to install one instance of the server and then create multiple domains and projects in order to organize your operations. I don't see a reason (based on your limited description) to install more than 1 QC Server.
My rule of thumb for separating between QC projects is mostly based on the sharing of testing assets, mainly tests and defects. I believe that if the tests can be reused by the ST and QA teams then you should create one project and organize the Test Lab in a manner that will make sense for your process.
If you are managing your defects in QC then you should also consider this as a factor to unite or separate the projects based on your customers for these defects.
In any case I would suggest starting with one and maximum separating the data later on since this is easier done than merging the data once the projects have been started.
[ QUOTE ]
System Test will be performed by IT Development prior to QA
[/ QUOTE ]
Will the development staff being doing total end to end testing or just unit and integration? If you're doing the automation, I don't imagine the developement staff is going to do a great deal of regression testing when they make changes.
Beantown Tester Go <font color="Red">Red Sox</font>