Take the 2-minute tour ×
Software Quality Assurance & Testing Stack Exchange is a question and answer site for software quality control experts, automation engineers, and software testers. It's 100% free, no registration required.

Which lightweight load and/or performance testing frameworks could you recommend?

some considerations:

  • Configuration with not too much coding (some JS would be okay)
  • command line + text editor (without complex IDE)
  • HTML/XML reports with simple graphics are just fine
  • it should be able to handle use scenarios (not just to load the main page of the site).

Just to start with what I've found:

http://perfcake.org

share|improve this question
    
What type of system are you going to be load testing? –  PaulDonny Mar 14 at 15:07
    
web application & web sites –  Alexei Vinogradov Mar 14 at 18:41

5 Answers 5

I have very good experience with the open source FunkLoad for load testing web-applications.

  • Has a test recorder (Uses proxy to record browser actions)
  • Generates Python code as test (You can edit it with a text editor)
  • Supports HTTP sessions and very complex scenario's (Login as users)
  • Scales up to a large number of threads per CPU
  • Has a great benchmark run which generates nice reports (also HTML)
  • Runs from the command-line (Easy to integrate in build servers)

I have used it to generate a thousand of simultaneous requests from a single computer. Also I created a test to randomly fill in 250.000 times a 50 pages long web-based questionnaires over the weekend on our new test production servers.

share|improve this answer

You can do a lot of this with JMeter, and it's free. The good thing about JMeter is that it is very commonly used and is also very powerful, so if/when you run into needing it to do more than the basics, it is ready for you to extend.

There is a GUI interface that you can use with no coding and get most basic scenarios up and going, and you can use either BeanShell scripting language or Java to extend it. I've had a lot of success with JMeter.

share|improve this answer
    
I found JMeter difficult to get started with, at-least the GUI part. Thats why I went looking for other alternatives. I think it does indeed cover the needs of the question. Seems I am not the only one who think JMeter is overly complex: jmeter-expert.blogspot.nl :) –  Niels van Reijmersdal Mar 14 at 20:32
    
It can be complex, but it's also very powerful :-). Coming from Visual Studio Load Test, a lot of the more light weight tools didn't give me what I was looking for, while Jmeter has (with writing some extensions to go along with it). There probably are better tools though for someone looking for something lightweight and simple. –  Sam Woods Mar 15 at 5:20

I'd go for JMeter as it's easy to start using.

It provides record-and-replay functionality via embedded HTTP Proxy Server or even from Google Chrome browser via plugin which is even better if you're behind proxy or recording SSL traffic.

It has thorough User Manual, large communities at stackoverflow, mailing lists, forums and tutorial video series which are the best place to start from.

share|improve this answer

Another one, I've found is Gatling

Probably - not the most lightweight, the scripts are written in Scala, but looks interesting.

share|improve this answer

As Ansi C is very lightweight, I recommend the latest version of LoadRunner, version 12. A 50 user non expiring version is available and feature-wise it buries tools like JMETER and Gatling. You could also check out SOASTA, a 100 thread non-expiring version is available at no charge. Also, feature wise, this is a commercial tool built for testing. SOASTA is a bit heavier on the per virtual user front as it uses JavaScript as it's core language.

Both tools have a long track record of success and defined capabilities in the CI space with Jenkins and similar.

If you're at the point where the financial risk is so large with failure that performance testing is now mandatory then what you will find when you measure the financial risk and compare it to the cost of any market tool that the tool cost is typically the size of a boil on the butt of a mosquito when compared to a risk the size of the cow the mosquito is feeding on.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.