2

For an application that is starting out I do not want to over pay for servers and resources that will not be used since the budget is very limited at this point. This will be a for an iOS and Android productivity app.

What is the best way I can accurately estimate the resources and grow them based on the number of users I get on so I never have crashes because the server is over worked.

I'm thinking of having 25% of my needs but don't know how to go about getting what are my minimum requirements.

Luis Liz
  • 131
  • 4

1 Answers1

3

You don't make it clear in your question, but I'm assuming you're talking about server side resources from a cloud vendor such as Azure, Amazon AWS, etc.

The answer is to start as small as possible to prove your app works and so you know you're on the "I'm not spending enough" side of the performance curve.

Do some load testing and ramp up the concurrent users until something breaks or response time becomes unacceptably long.

Increase your server resources (either up to bigger machine(s) or out to a bunch of machines in parallel) and repeat the load testing as necessary.

There's no magic number (e.g. "25%") that will solve your problem for you - you'll just have to experiment and find the balance that's right for you between performance, available reserve, and cost.

Dan Pichelman
  • 13,773
  • 8
  • 42
  • 73
  • @LuisLiz: This question is too broad, but this answer is as close as you are going to get. The only other thing I would add is have two of your own computers. One to run the services, and another to execute some load tests against that one computer. How many users does it support? That at least gives you a baseline for almost free if you have a second computer. – Greg Burghardt Nov 22 '19 at 17:05
  • Yeah, now I noticed that it can seem broad. I was thinking in terms of estimates but after thinking reading the answer and thinking about it there's no other way without experimenting I was really having the wrong approach. – Luis Liz Nov 23 '19 at 15:57