Testing at scale


I’d like to run a test of a lot of users (50,000 or more) executing one specific action in an app at the same time.

Do you know any automated testing platforms that can simulate 50,000 users executing an action in a bubble io app at the same time?

The main objectives of the test are:

  1. Check how many simultaneous users running the action would make the app crash (if it crashes).
  2. Check the server capacity use.

Please don’t hesitate to share any solutions that you know for testing Bubble io apps at scale, even if they doesn’t seem like a good fit for what I proposed.

Thank you!

I think a smaller number to start makes more sense, but beyond that, it entirely depends on what your app is and how it functions. What are the points of potential limitations and such.

Apps like Reddit and Twitter are likely not heavily dependent on cpu, but rather fast db use. But then you can have other apps that are much more computationally intensive that would require another way to test.

So perhaps give us the rundown of your app and people can come up with suggestions?

For some simple tests you can use https://loader.io. A little more advanced but more capable is https://www.flood.io. Testing applications with load seems to be very expensive from what I’ve seen.

Personally I use loader.io more often and I create a page that has some contrived workflows that run on it to test what it is that I want. I don’t test real user flows (not that they’re bad–I just don’t have the development capacity to build those out). You could also test a handful and do the math to approximate how well that scales.

Hey , this is the feature that I want to test. It is a bit complex to explain all the workflows and constraints here but I’ll try to summarize:

Basically, it is a search engine for university courses. You select country, then a city inside of this country, then an educational institution inside of this city (if you don’t find it you can create it), then a study program inside of this institution (if you don’t find it you can create it), then a course inside of this study program (if you don’t find it you can create it).

Every time a user does a search, those search paremeters are saved and they would be loaded automatically next time, except for the course field.

This is a very summarized explanation, there are a lot of workflows and db searches setup to get this done.

@flowtron this tools seem interesting, thanks!

This topic was automatically closed after 70 days. New replies are no longer allowed.