I am collecting cpu utilization data for ec2 instances using prometheus.
For each data entry, it includes instance_id, and cluster_id and some other
information. A cluster can have multiple instances, and the # of instances
vary due to scaling up and down.
Now I want to create a dashboard for following:
Show the cpu utilization data for all cluster with # of instances >= 50,
and with avg cpu utilization <= 10%.
To do this, I first need to select all the cluster with avg # of instances
>= 50. I used following query to select these clusters:
count by (cluster_id)(avg_over_time($Metric{}[1200s])) >= 50
I am following this blog (
https://www.robustperception.io/graph-top-n-time-series-in-grafana ) to do
this selection.
When using the avg_over_time function above, I found it’s not clear to me
what this function is actually doing. From its definition:
-
max_over_time(range-vector): the maximum value of all points in the
specified interval.
So couple questions:
1.
My understanding is that, my raw data is a time series, and after
applying this max_over_time function, the result will be another time
series, is this right?
2.
Which interval to use? And its impact of the result?
I used 1200s in above query, instead of the $_range_s suggested in the
above blog. I used 1200s because I noticed it can give me some reasonable
result comparing to $_range_s. But again, I don’t know how this parameters
affecting the selecting result. Anyone can help to understand?
1.
What’s the time range this query apply on? Does it automatically apply
to all the available date, for example, if my data is 1 year long, does it
apply this query for all the data points in past year? Can I configure it
to be like last 7 days?
2.
Finally, is my above query correct? Will it give me all the cluster with
# of instances >= 50? I want to verify this but I don’t find a good way to
validate.
Can someone help with my questions? Thanks!
--
You received this message because you are subscribed to the Google Groups
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/6be28a51-7404-4661-a7c5-67cf9143fe7en%40googlegroups.com.
