The basics of Threads and Test Delays
Hi
I am new to SOAPUI and would appreciate if someone could iron out a few queries I have. I'm using SOAPUI (Not PRO) to create load tests and have read the online information but it's not quite as clear as I'd like. Essentiall what I want to achieve is the following:
A Suite of tests which load test services (initially I need a baseline) then I'll move onto more complex load testing. I understand 'Threads' to be concurrent requests to the service under test (even though it's described as 'virtual users'). Am I right in this assumption?
Assuming I am, I want to ramp up the Threads to high numbers to test the average response time at different thread values.
I'm using the 'simple' strategy and my understanding was that if I have say 10 threads, then 10 concurrent requests will be sent to the service and then another 10 after the 'delay time' and then another 10 after the 'delay time' and so on.
But reading the documentation it appears to suggest that the delay time is the time between the end of the initial 10 and the beginning of the 2nd 'batch of 10', which is different.
So which is it?
If it's the latter, then whats the reason for the delay at all since everything has finished before that delay even starts so surely it's ok to send another 10??
Secondly, how do I extract all of the data that has been generated? When I export the stats using the icon at the top it just gives me the data presented on screen. Can I get at the more granular data, the timings for each request? I want to build up a graph of maybe several thousand individual timings using say 10 threads and then a similar graph using 20 threads etc etc
Thirdly, are there not any scales or labels on the axis for the graphs produced?
I am new to SOAPUI and would appreciate if someone could iron out a few queries I have. I'm using SOAPUI (Not PRO) to create load tests and have read the online information but it's not quite as clear as I'd like. Essentiall what I want to achieve is the following:
A Suite of tests which load test services (initially I need a baseline) then I'll move onto more complex load testing. I understand 'Threads' to be concurrent requests to the service under test (even though it's described as 'virtual users'). Am I right in this assumption?
Assuming I am, I want to ramp up the Threads to high numbers to test the average response time at different thread values.
I'm using the 'simple' strategy and my understanding was that if I have say 10 threads, then 10 concurrent requests will be sent to the service and then another 10 after the 'delay time' and then another 10 after the 'delay time' and so on.
But reading the documentation it appears to suggest that the delay time is the time between the end of the initial 10 and the beginning of the 2nd 'batch of 10', which is different.
So which is it?
If it's the latter, then whats the reason for the delay at all since everything has finished before that delay even starts so surely it's ok to send another 10??
Secondly, how do I extract all of the data that has been generated? When I export the stats using the icon at the top it just gives me the data presented on screen. Can I get at the more granular data, the timings for each request? I want to build up a graph of maybe several thousand individual timings using say 10 threads and then a similar graph using 20 threads etc etc
Thirdly, are there not any scales or labels on the axis for the graphs produced?
