Re: redo curiosity
Date: Thu, 18 Feb 2010 13:24:14 -0500
does one database happen to have forced logging?
The other thing is (i'm making an educated guess here) is that more info is being sent to redo logs in newer versions, especially for things like log miner.
Joe Testa, Oracle Certified Professional Senior Engineering & Administration Lead
Interested in helping out your marriage? Ask me about "Weekend to Remember"
Maureen English <maureen.english_at_alaska.edu> To:
"Oracle-l_at_freelists.org" <Oracle-l_at_freelists.org> Date:
02/18/2010 01:18 PM
We have an Oracle 8i database and an Oracle 10g database that currently both serve the same purpose. We're moving off the 8i to the 10g, just not as quickly as we'd hoped.
The two databases have the same accounts and the same jobs running against them...both are reporting instance databases, so users do very little in terms of creating and updating anything. The 2 databases have the same materialized views with the same nightly refreshes done at the same time each night...if the materialized view in one database is done with a fast refresh, that's how it's done in the other database.
Both databases have the same size redo logs, but the 10g database ends up with 2-3 times as many archive logs as the 8i database.
I'm looking at the database parameters to see what's so different...any other suggestions as to why the 10g database would be generating so much more redo?