RE: RE: checkpoint incomplete issue

If we approach the topic from as high a level as possible wouldn't we want all of the links in the chain to be as fast as possible. Should it matter if my data is on disk, in memory, stored as light molecules...the problems occur when the fast parts upset the balance of power. We end up ignoring SQL statements that would have taken 2+ years on an older system but now resolve in < 2 minutes. 6 months down the road our 8 * 2Ghz processes can't keep up anymore and management is looking at a 24 way. Hardware vendors love it, heck I bet Oracle loves it too.

-----Original Message-----

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Jeremiah Wilton Sent: Friday, November 19, 2004 12:38 PM To: oracle-l_at_freelists.org
Subject: RE: RE: checkpoint incomplete issue

Hello folks. Long long ago in the days of slow disk arrays and filesystems that serialized file access, I wrote a blurb on "checkpoint not complete."

It is here:

http://www.speakeasy.org/~jwilton/oracle/checkpoint-not-complete.html

Please enjoy.

BTW, I would enjoy a debate on the question of "how many caches do we need to run Oracle?"

Oracle Buffer Cache
UBC/filesystem cache
Async write queue
Disk array hardware cache

More? less? Will data ever be on disk or is the database just imaginary?

--

Jeremiah Wilton
Independent Oracle Professional
Oracle Certified Master
Disaster Recovery - Seminars - Technical Interviews http://www.speakeasy.net/~jwilton
--

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

http://www.freelists.org/webpage/oracle-l Received on Fri Nov 19 2004 - 13:45:33 CST