[cloudwatch exporter] API usage optimization

60 views
Skip to first unread message

Or Shachar

unread,
Apr 19, 2022, 11:16:48 AM4/19/22
to Prometheus Developers
Hi,

Our company recently hit this issue - with `/metrics` endpoint taking > 20s to load.
I started working on a solution and I saw that big contributions should start with a short discussion here.
Pardon for starting the work in advance. 

I'd really appreciate a short review of the suggested solution.

Also - I wonder if we should migrate to https://github.com/nerdswords/yet-another-cloudwatch-exporter even though it's not under the official Prometheus organization.

Thanks in advance!

keywords:
- getMetricStatistics
- getMericData
- cloudwatch

Stuart Clark

unread,
Apr 19, 2022, 12:02:22 PM4/19/22
to Or Shachar, Prometheus Developers
On 2022-04-19 16:14, Or Shachar wrote:
> Hi,
>
> Our company recently hit this issue [1] - with `/metrics` endpoint
So are you intending on trying to move to using the bulk multi-request
API call where possible to reduce the number of calls & make things
quicker/cheaper?

--
Stuart Clark

Or Shachar

unread,
Apr 20, 2022, 7:51:19 AM4/20/22
to Stuart Clark, Prometheus Developers
Not sure about cheaper because the pricing is the same (per metric requested).
But for sure it will be much faster. One request of GetMetricData can retrieve the same amount of data as 500 requests of GetMetricStatistics.
--
בברכה,
  אור שחר
Reply all
Reply to author
Forward
0 new messages