I understand that TS is using direct HTML code to server instead of using a 'standard scripted browser playback' approach. I can see the pro/cons for each. One limitation that I see is that the direct html method might not capture nor reproduce the same effect...
so my question is how does TS - Load work in this example. Will it only measure and apply user load to Server A, or A and B?
in this scenario - Server A and Server B are not on the same network.
- A User keys in and searches for data based on parameters using a Web application front end in server 'A'
- Then web application sends that data to another application running on server 'B' which compiles the information
- compiled data is then housed in server 'B' and the web application running server 'A' is sent an acknowledgment screen
- User then either excepts or rejects the acknowledgment screen, if rejected restart from beginning; if accepted server 'A' tells server 'B' to generate the data based on all parameters stored in server 'B'.
- Once server 'B' completes the process an acknowledgment page is sent back to server 'A' and user can then push the data to server 'C'
1 Answer, 1 is accepted
In short, yes Test Studio's load test should give you the desired results you're looking for. Technically our load test is sending direct HTTP requests (not HTML only because HTML is the data contained in HTTP requests) to your server/servers en mass. When you create a User Profile you are actually recording any/all HTTP traffic that is flowing to/from your browser of choice. If the browser is talking to multiple servers (e.g. JavaScript might be making web service calls for extra data to something other than server A) that extra traffic will be recorded and added to the user profile. If all HTTP traffic is to Server A, and server A talks to server B and C, it will work the same in a load test. All the servers will not be able to tell the difference between a real browser and a Test Studio load test.
Kind regards,Cody
the Telerik team
Test Studio Trainings