Last week, we held one of our most popular webinars on “Managing Continuous Performance in Your CI/CD Pipeline.” During the 60-minute session, we covered why continuous testing is important, how to conduct a risk analysis to determine test coverage, and how to integrate load testing into Jenkins and manage these tests in a tool like Zephyr.
We received a ton of questions during the session, and it became clear that people wanted to learn more about LoadNinja, its capabilities, and overall best practices for performance testing. So, we put together this FAQ blog to answer all your questions!
Regarding best practices, is it better to use large tests with more complicated scenarios (with lots of actions, pages, etc.) or with simpler scenarios? Are there pros and cons?
At the end of the day, it’s up to you. The overall goal of performance testing is to model end user behavior as closely as possible and to analyze the resulting data. While a single user may go through a long, complicated use case, it may not be worth the marginal cost associated with the time and effort needed to record these actions.
Can we use a single VM for 10,000 VUsers?
Since LoadNinja runs on real web browsers, we wouldn’t necessarily want 10,000 users coming from a single load generating agent.
By using real browsers, LoadNinja creates the most realistic and accurate representation of load on the infrastructure supporting the web application under test.
Read more about browser-based load testing.
Can we use LoadNinja in Dev and Test environments (inside firewall)?
Yes, you can use LoadNinja in Dev and Test environments to test inside firewall. You can do this by downloading the LoadNinja endpoint and selecting “internal” when entering the URL for the website.
Check out our documentation on how to test web apps in local network with endpoint.
Is it possible to parameterized tests to send different users’ credentials, for example?
Yes, this is possible and can be done by adding a databank. To create a data-driven load test in LoadNinja, attach databanks to individual scripts that your load scenario will run. Then, map databank columns to event parameters.
Can users update selectors through AI for the whole application?
Yes, clicking the gear icon within the recording will display the global Playback Settings applicable to the entire web test. From there, you can select the object identification used and the change will apply to all events in the recorded web test.
Read more about our AI Objection Recognition and Instaplay Recorder.
Can you distribute users over time?
Yes! When creating the load test scenario, you can use the “ramp up time” and “ramp up steps” options to distribute the increase of users over time.
You can also distribute the load between various locations. If your users are global, you can check the accessibility of the tested website or application throughout the world.
Read about LoadNinja’s geo load distribution.
How well does LoadNinja work within a Multi-Brank Pipeline? Any major gotchas?
LoadNinja should be able to work with Multi-Branch Pipelines and execute without any major gotchas!
What is the difference between TestComplete and LoadNinja? It seems that both tools can test browser applications.
LoadNinja is used to test load on Web applications while TestComplete is used for functional UI testing for one user, and does not have the ability to simulate thousands of virtual users.
Do you have a published list of API calls so that I could integrate with Azure DevOps?
Yes, you can find the LoadNinja Public API here.
If you would like to learn more about LoadNinja, check out our features page! Or if you’re ready, you can start your trial now!