I would like to know if it is possible to evaluate and optimize AWS spending (specifically, EC2 spending) without having access to account billing info?
Long story short, we do not have the ability to view account billing dashboards / metrics due to a company policy. (Such a permission would allow us to see other billing metrics that are sensitive).
However, I can get a monthly snapshot of our AWS bill, with a breakdown for each service.
I know that our EC2 spending is very high. But without having the ability to drill into this data to get more insights, how do I optimize? I have some theories on where most of our spending is coming from (EMR/spark compute nodes), but I can't know for sure unless I can measure it.
Is there a way for me to get compute hours usage for EC2, without going through billing dashboards / APIs? I imagine if I can get the total hours for a machine, and multiply it by the hourly rate, I should be able to generate a cost estimate myself?
RE: close vote:
I think this is not off topic. I am asking for a way to programmatically measure the compute hours for AWS virtual machines (EC2). This seems to be in scope for stackoverflow.