Greg,

Be careful with that. We always tend to assume that because the top
wait event is cpu, that the cpu
is busy. Perhaps you have alot of processes in the blocked or waiting
state .... this could bind things
up as well. Many times, folks put all kinds of low impact ;-)
applications like application server middle
tiers and performance monitors that should not affect performance. I'm
not saying that databases are
not resource pigs .... they are, but they are not the root of all evil
either. I've even seen CPUs at 98%
idle waiting for processes to complete. vmstat could be your friend in
this case. Because you are on
a Sun, lockstat will help alot as well.

Thanks!

Jack

greg wrote:

>DBAs,
>
>I've got a 9.2.0.7 instance on a Sun E450 where response times have been
>getting steadily slower over the last 4 weeks.
>
>Statspack says my top wait event is CPU time which has risen from 80% to a
>whopping 92% during this time. I thought I needed more processor power but
>Solaris is showing 40 - 50% idle time.
>
>So what could be driving my CPU wait time so high and slowing down my
>response time?
>
>Thanks,
>Greg
>
>
>---------------------------------------------------------------------
>TO REPLY TO EVERYBODY , PLEASE CLICK REPLY-ALL, NOT JUST REPLY
>To post a dba job: http://jobs.lazydba.com
>To Subscribe : http://www.LazyDBA.com
>To unsubscribe: http://www.lazydba.com/unsubscribe.html
>
>
>
>
>


Oracle LazyDBA home page