RE: LGWR using lots of CPU time, low CPU usage

The ultimate sincerest form of flattery is for someone to attribute something smart to you that you wish you had done but, alas, did not actually do.

(It was Tim Gorman who posted the excellent analogy.)

Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com

Upcoming events:

- Hotsos Clinic, Dec 9-11 Honolulu
- Hotsos Clinic 101, Jan 7-9 Knoxville
- Steve Adams's Miracle Master Class, Jan 13-15 Copenhagen
- 2003 Hotsos Symposium, Feb 9-12 Dallas

-----Original Message-----
Jared.Still_at_radisys.com
Sent: Tuesday, November 26, 2002 4:07 PM To: Multiple recipients of list ORACLE-L

> >And "old school" is still right about not putting RedoLogs onto
RAID5.
> From what I'm being told, this is not your father's RAID5. This is
what

>they tell me:

> The CPU hands the IO to the disk controller and rather than do the
> physical disk IO while the process waits, the disk controller caches
> it to local memory and says done. Therefore, effectively there is no
> wait for IO and it doesn't matter if we are RAID 5 or RAID 0+1,
> the system is NOT waiting for the IO. He said the only time there
might
> be a delay is during the cache's battery refresh times. I checked your
> dates and it was not occurring during those times. Also, if you look
> at the iostat statistics under the 'wait' and '%w' headers you will
> see all zeros.

Debi,

That is true, up to a point.

Think of the cache as a water tank. You have a garden hose filling up the tank. You can keep increasing the water pressure for a while.

But the outlet at the other end of the tank has a fixed capacity. It flows 10 GPM, and no more.

What happens when you increase the flow at the intake to 20 GPM? The tank fills up.

When the tank fills up, your intake flow will need to decrease, because you can only flow 10 GPM at the outlet.

Now, think of the outlet as writing to disk, the RAID5 cache is the water tank, and your database is the inlet that wants to run at 20 GPM.

If your database activity will never be intensive enough to stress the cache like this, no problem. But 'never' is a very long time.

If any of this sound familiar, Cary Millsap posted a very similar explanation a few weeks ago.

Plagierism is the sincerest form of flattery. :)

Jared

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: 
  INET: Jared.Still_at_radisys.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Cary Millsap
  INET: cary.millsap_at_hotsos.com

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).