Dashboards & Visualizations

pctLoad, pctUser, pctNice etc?

sarnagar
Contributor

Hi,
What do these keyword indicate pctLoad, pctUser, pctNice etc?
For ex in the below query what does "as pctLoad by host" refer to?
index=sys_* sourcetype=vmstat host=sl73bsaapd001* | multikv fields memFreePct | timechart span=1h avg(memFreePct) as pctLoad by host

Tags (3)
0 Karma
1 Solution

Richfez
SplunkTrust
SplunkTrust

They are percent "things". Percent User (pctUser) is the portion of CPU time spent on User tasks (as opposed to "system" tasks)... pctNice is how much CPU was taken up by processes that were prioritized lower than usual, and so on.

This answer at serverfault seems short and reasonably explanatory for most of those.

That leaves "pctLoad". As you can see from the search you posted, pctLoad is converted from an average of memFreePct, which is how much memory the system has free. I'm not 100% sure which number memFreePct is (it could be all memory minus memory allocated, all memory minus memory allocated but ignoring cache or a couple of other basically the same but slightly different metrics), but in any case it's a basic "how much RAM do I have free" indicator. When that gets very low it is usually considered not as good as when it's not very low. (Though on the flip side, if you have too much free perhaps you simply have too much RAM in that machine for its workload and it's just sitting there not being useful).

View solution in original post

0 Karma

Richfez
SplunkTrust
SplunkTrust

They are percent "things". Percent User (pctUser) is the portion of CPU time spent on User tasks (as opposed to "system" tasks)... pctNice is how much CPU was taken up by processes that were prioritized lower than usual, and so on.

This answer at serverfault seems short and reasonably explanatory for most of those.

That leaves "pctLoad". As you can see from the search you posted, pctLoad is converted from an average of memFreePct, which is how much memory the system has free. I'm not 100% sure which number memFreePct is (it could be all memory minus memory allocated, all memory minus memory allocated but ignoring cache or a couple of other basically the same but slightly different metrics), but in any case it's a basic "how much RAM do I have free" indicator. When that gets very low it is usually considered not as good as when it's not very low. (Though on the flip side, if you have too much free perhaps you simply have too much RAM in that machine for its workload and it's just sitting there not being useful).

0 Karma
Get Updates on the Splunk Community!

How I Instrumented a Rust Application Without Knowing Rust

As a technical writer, I often have to edit or create code snippets for Splunk's distributions of ...

Splunk Community Platform Survey

Hey Splunk Community, Starting today, the community platform may prompt you to participate in a survey. The ...

Observability Highlights | November 2022 Newsletter

 November 2022Observability CloudEnd Of Support Extension for SignalFx Smart AgentSplunk is extending the End ...