Closed kushalShukla-web closed 1 week ago
Solves #705
Hi, I commented on a few, but they are all likely problemmatic. I recommend you change tack to use metrics exposed by the fake webserver, since there are a lot of them and prombench can control them.
Actually i have tested this on one of the Pull Request where the prombench was running.
Ok but my recommendation remains the same.
I have replaced some of the existing metrics with new ones from PromBench, such as:
How many of those did you count?
Hi @bboreham all the queries having value above 2000 , and codelabz metric is having 52000 different variations .
I trimmed down the newly-added queries a bit:
_bucket
series directly; those should be used by histogram_quantile
or similar.k
parameters to topk
.I also trimmed down some pre-existing queries for balance, to retain about the same overall load on the server as before.
This PR enhances the synthetic load generation by incorporating additional PromQL operators in the 6_loadgen.yaml file:
topk
function to test query performance with ranked results.