Is there any way for us to perform load testing on Backendless to see how much simultaneous/real-time/concurrent users and API calls it can actually handle? Thanks.
It can be as simple as this:
- Write a testing app that makes calls
- Run it
For example, I would like to test whether the Cloud9 plan actually has 10,000,000 API calls limit per month. Testing API calls capacity is easy because I can do like as what you have said, by writing an app that call to a specific API endpoint I created in the console 10,000,000 times to max out the limit, or use any cloud-based tool out there.
However, for testing simultaneous connections/real-time connections in Cloud9 plan, that means I have to use 100,000 unique (virtual) users that connect to Backendless, am I right?
Cloud9 is not limited 10,000,000 calls. That’s the amount that’s included in the plan. As for the strategy to simulate concurrent users, I will leave it up to. You should be able to do it with either virtual or physical users. A more important question is why? If you doubt we can do what we promise, do you think we’d able to get where we are?
Wow, I think you are a little bit harsh there. As I said before, it’s just for studying, I don’t doubt the platform capability. I am currently trying to implement ISO/IEC standard-based quality evaluation on my app and I need to evaluate as many aspects as possible from my app. That’s why. I just following what the ISO/IEC standard has defined for measuring performance of an app, in which one of them is about user access capacity and API calls/transaction capacity.
After all it’s impossible to actually measure the API calls capacity because even though the plan is throttled to 10,000,000 calls/month, it’s actually very scalable to unlimited calls for some fee, right?
I think the specification is very clear. There’s no need for evaluating this metric.
I would not call it “throttled”. 10,000,000 is a block of calls included with the plan. And yes, as you described, the calls are unlimited for an additional fee. Here’s what the pricing table on the website shows:
