You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
I'm using Loki to gather Netflow data. The unwrapped value is a "Bytes" counter, which is always increasing. I need to calculate "top talkers" and get the total data transferred. Right now I'm doing last_over_time()-first_over_time(), which works, but it's a little slow since there are thousands of unique label sets/conversations.
Also, last_over_time()-first_over_time() won't properly deal with breaks in monotonicity like increase() would/should.
Describe the solution you'd like
Prometheus has an increase() function for counters. It would be nice to be able to use that, I assume it would be faster than last_over_time-first_over_time, since it isn't doing 2 different queries?
Describe alternatives you've considered last_over_time()-first_over_time()
Additional context
This is what I'm doing currently to get the top 20 conversations (this is done in Grafana so there are some dashboard variables in use):
Is your feature request related to a problem? Please describe.
I'm using Loki to gather Netflow data. The unwrapped value is a "Bytes" counter, which is always increasing. I need to calculate "top talkers" and get the total data transferred. Right now I'm doing
last_over_time()-first_over_time()
, which works, but it's a little slow since there are thousands of unique label sets/conversations.Also,
last_over_time()-first_over_time()
won't properly deal with breaks in monotonicity like increase() would/should.Describe the solution you'd like
Prometheus has an increase() function for counters. It would be nice to be able to use that, I assume it would be faster than last_over_time-first_over_time, since it isn't doing 2 different queries?
Describe alternatives you've considered
last_over_time()-first_over_time()
Additional context
This is what I'm doing currently to get the top 20 conversations (this is done in Grafana so there are some dashboard variables in use):
The text was updated successfully, but these errors were encountered: