Sorry for the questions answered with still more questions, but.... What version of Oracle? Have you considered prebuilding the snapshot via export/import and offline instantiation? This is possible if you pre-create the snapshot log and then fast refresh to catch up after the build. Look up "offline instantiation" in metalink. Do you really feel as though the 5.7g resource requirement is all that unreasonable? I have to think that the sorting for the index is causing it. We have several 2g datafiles for each of our TEMP and ROLLBACK requirements in 8.1.7 databases, and they often autoextend > 20g. We then resize them as needed. Finally, I must ask how and why you need 120+ database instances in this day and age. All of the overhead associated with each should be considered a waste of some of these same resources we are discussing here. Things such as TEMP ROLLBACK and SYSTEM overhead redundancy should be eliminated where possible before serious reductionary steps are considered. I do realize sometimes these things are not completely within our control, but the time when such issues come up is great to mention it! Michael Fontana Sr. DBA NTT/Verio -----Original Message----- From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Stephen.Lee@xxxxxxxx Sent: Thursday, April 15, 2004 11:13 AM To: oracle-l@xxxxxxxxxxxxx Subject: Fat Pig Replication ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------