Member-only story
Is Bigger Really Better? Part 1
A study on cloud instance size, performance, and cost.
The Premise
As a cloud architect one of the things you have to constantly be aware of is your cloud spend. How much you’re spending and what your budget is has a very real impact on a number of different areas, including what kind of performance you can expect out of your infrastructure and, by extension, your applications. Designing an architecture that not only fits your monetary budget but also your performance budget and is something that is easy to manage, operate, and troubleshoot can be a tall order at times. The best architects find a way to balance all of these factors; the worst do the opposite. But is there really a direct correlation between cost and performance? Is there a way to spend less and get better performance out of our clouds? Maybe.
The Hypothesis
This is where things start to get interesting. First we need to make a few assumptions:
- You have a Python 3 application that you need to deploy. Its a typical stateless 12-factor application that takes HTTP API calls and does some stuff in a database elsewhere. Nothing fancy but fairly typical.
- Your cloud is AWS.
- Your maximum payload is 1MB.
