Welcome to the new Parasoft forums! We hope you will enjoy the site and try out some of the new features, like sharing an idea you may have for one of our products or following a category.
Traffic Shaping Question in LoadTest Tools
LegacyForum
Posts: 1,664 ✭✭
Hey guys. Is there a way to apply the global Hits/sec or Num of Users traffice schedulintg to individual test cases? Running test cases within test suites (folders) concurrently is a good thing, but i want to be able to ensure that each test case I run executes for either the time allotted or the number of requests alloted. Rather than having some service operations fall short due to resource sharing or round robins misses.
It would be really cool to be able to assign number of requests independently to each test case. For instance, Test Case 1 exectues 100 request. Test Case 2 executes 50 requests etc...
Thanks in advance,
LeapTester
It would be really cool to be able to assign number of requests independently to each test case. For instance, Test Case 1 exectues 100 request. Test Case 2 executes 50 requests etc...
Thanks in advance,
LeapTester
0
Comments
-
There are two separate things to consider in order to setup what you have described. First, are you familiar with the mechanism for setting up ratios between virtual user scheduling? In the load test scenarios view, there is a tab for "Users" which allows you to set the total number of users. If you have more than one virtual user profile defined, you will also see a "Profiles" tab where you can specify the distribution of the virtual users across the total number scheduled. In particular, if you schedule 150 users, and you create a 2:1 ratio between two profiles, you would have 100 of one type of user and 50 of the other type of user at any given time. This, of course, is only possible if there is no data dependency between the two test cases. If input for Test 2 depends on response data from Test 1, then Test 1 will have to run as many times as Test 2 does. If there are no dependencies, then each test case can be placed in its own test suite, and virtual user profiles can be created and mapped onto the two test suites.
Secondly, are you familiar with load test stop actions? You can define a script which will halt the load test based on a given condition. So you could, for example, halt the load test once a particular test case runs 100 times, rather than having the load test run for a particular duration. If this sounds promising to you, we can supply you with some examples on how to set this up.
Keep in mind that in both methods, there is imprecision. If you specify a stop action at 100 hits, it will simply signal the load tester to halt, however there may be additional processes that may need to continue (such as iterating through all rows of the data source).
Is this sufficient for what you had in mind?0 -
Hey TomT. Thanks for the response.
Let's assume I know nothing With that said, what you've described will be great for what we need to do. However I fear the setup time and the the interpretation of the results might be a more than i'm willing to tackle right now. At anyrate lets do it anyway .
I am familiar with the basic usage of the load test clients, profiles, machines and only marginaly familiar with ratio setup and the "stop on" criteria methods. I can re-read the manual to get better at that. However i'll probably be bugging the forum periodically for assistance with this.
Thanks again,
LeapTester0