I once spent several days trying to figure out the information from the
dba_hist* tables. I never really managed to get a calculation I was happy
with, but you can start with dba_hist_sysmetric_summary, look at 'CPU Usage
Per Sec' and 'Background CPU Usage Per Sec'. You get entries for each
snapshot, so you can get a decent idea of what oracle sees. But I really
have no good idea how accurate that is going to be. Especially with
multiple db instances on a server.
On Thu, Sep 14, 2017 at 8:52 AM, Dominic Brooks <dombrooks@xxxxxxxxxxx>
wrote:
Yep.
Plus:
3) The connection pools are way over-sized – see real world performance
group videos
*From: *Mladen Gogala <gogala.mladen@xxxxxxxxx>
*Sent: *14 September 2017 12:49
*To: *dombrooks@xxxxxxxxxxx
*Cc: *peter.fong@xxxxxxxxx; Chris Taylor
<christopherdtaylor1994@xxxxxxxxx>; ORACLE-L <oracle-l@xxxxxxxxxxxxx>
*Subject: *Re: Oracle CPU Time Used, vs CPU Time Available (CPU Capacity)
On Thu, 14 Sep 2017 07:01:17 +0000
Dominic Brooks <dombrooks@xxxxxxxxxxx> wrote:
I have spent a lot of effort recently explaining that running at asustained 95%
utilisation over a 1-2 hour period does not just mean that you’ve got anextra 5%
that you could be squeezing out...
In case you described,you probably do not have that capacity. 95% of CPU
usually means run queue and kernel overhead which will easily devour the
remaining 5%. Such sustained consumption can mean one of two things:
1) You need a bigger box (or vbox).
2) There is a sub-optimal task running on the machine which needs to be
optimized.
--
Mladen Gogala
Oracle DBA
Tel: (347) 321-1217
