Re: Enetrprise Manager CPU Sampling and Mesurement

From: Karl Arao <karlarao_at_gmail.com>
Date: Tue, 12 Nov 2013 17:11:45 -0500
Message-ID: <CACNsJnfvRfps0e9+6vVoU9EEb1ngAEBezrdeKaZjJp4yepogbw_at_mail.gmail.com>

The collection is every 15 minutes, I made use of the following:

$ cat metric_current.sql
col tm format a20

col target_name format a20
col metric_label format a20
col metric_column format a20

col value format a20
select TO_CHAR(collection_timestamp,'MM/DD/YY HH24:MI:SS') tm, target_name, metric_label, metric_column, value
from mgmt$metric_current
where metric_label = 'Load'
and metric_column = 'cpuUtil'
and target_name = 'desktopserver.local';

oracle_at_desktopserver.local:/home/oracle/dba/karao/scripts:emrep $ cat metric_currentloop
#!/bin/bash

while :; do
sqlplus "/ as sysdba" <<! &
spool metric_current.txt append
set lines 300
_at_metric_current.sql
exit
!
sleep 10
echo
done

while : ; do cat metric_current.txt | awk '{print $2}' | sort | uniq ; echo "---"; sleep 2; done

---

15:11:59
15:26:59
15:41:59
15:56:59

Yes they are being rolled up to hourly and daily, in OEM12c there's a new view called GC$METRIC_VALUES_DAILY where the Consolidation Planner pulls the OS utilization data.
I've got some notes on the following link http://goo.gl/Tv2uh4

-Karl

On Tue, Nov 12, 2013 at 9:10 AM, Ls Cheng <exriscer_at_gmail.com> wrote:

> Hi Niall
>
> Thanks for the reply.
>
> I dont have anyrepository handy so I cannot query MGMT_METRICS_* data so
> may i ask you if the raw data is sampled every minute? And when the raw
> data is aggregated per day does it mean it adds the CPU utilization during
> 24 hours and get the average?
>
>
> Thanks
>
>
>
>
>
>

-- 
Karl Arao
Blog: karlarao.wordpress.com
Wiki: karlarao.tiddlyspot.com
Twitter: _at_karlarao <http://twitter.com/karlarao>

--
http://www.freelists.org/webpage/oracle-l

Received on Tue Nov 12 2013 - 23:11:45 CET