You use the Load Type setting along with Load Profile to specify how your test should simulate the load. You can use either VUs or Rate load type for your tests.
When you selected the VUs type for a test, the load profile settings specify the number of virtual users working on the server simultaneously. This number can be permanent (fixed) during the entire test run, or it can change according to the selected load profile.
If some virtual user stops working because its TestCase is over, LoadUI waits for the specified Wait Time and then starts a new user to maintain the needed number of virtual users. The delay before restarting the virtual user simulates the pause between one user leaving the server and another user starting to use it.
|We recommend that you have some wait time for your requests. If you set the 0 wait time and the target simulation takes a very short time to be executed, LoadUI may require more processing power that a computer can provide.|
LoadUI keeps the specified number of virtual users even if the server slows down.
For non-fixed load profiles, it is quite possible that at certain points of time the current number of virtual users will differ from the number specified by the profile. In this case:
If the current number is less than the needed number, LoadUI will start extra virtual users to reach the needed number.
If the current number is greater than the needed number, LoadUI will not terminate "extra" virtual users. It will simply wait until their simulation is over.
If you selected the Rate load type for a test, the load profile settings specify how often "users" visit the tested server. That is, the profile does not specify the number of concurrent virtual users on the server, but specifies the number of users arriving to the server every second, minute or hour.
If some virtual users stop, LoadUI does not restart them.
If virtual user simulation takes longer than the arriving period, a severe increase in the server load may occur.