Peter, Haven't had much luck with reducing the amount of data. Too often it turned out there were problems with missing data, and the like. For example, we saved space by not copying all the historical payroll data. The upgrade testing failed to catch a data situation that wasn't represented in our data subset. The more reliable methods I've used are things like recovering a backup. This has a bonus of actually getting confidence in your backups. Export/import offers many creative alternatives, but is only good for smaller amounts of data. The big issue with reducing DBA time is often the negotiation. One group wants the data refreshed, another group isn't ready. Somebody creates or modifies database objects or data and your refresh wipes them out. So then they scream (to your manager) to never refresh the data unless you check with them first. Sometimes the DBA time savings is only refreshing on demand. Dennis Williams On 6/18/07, Peter Barnett <regdba@xxxxxxxxx> wrote:
Thanks for the thoughts. We are likely going to be refreshing and rebuilding at least three multi-terabyte databases every month for development. Same for systems and integration testing. Speed is definitely an issue. We would also like to reduce the number of DBA resources required to do this. That is what originally drove the Symclone question. Can we do this with little, or no, DBA time? There seems to be agreement that it is slow. Guess we need a 'Plan B'. --- Matthew Zito <mzito@xxxxxxxxxxx> wrote: > And actually, as I think about it, if you're just > looking to move data > from one place to another, the symclone stuff might > be a bit of overkill > for it - if you just want to move a chunk of data > from one place to > another, you could use a timefinder snapshot, which > would reduce your > disk usage, or if its done infrequently, you could > use a standard BCV, > which provides better performance than the symclone > piece. > > Thanks, > Matt > > -- > Matthew Zito > Chief Scientist > GridApp Systems > P: 646-452-4090 > mzito@xxxxxxxxxxx > > > -----Original Message----- > > From: oracle-l-bounce@xxxxxxxxxxxxx > > [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf > Of Peter Barnett > > Sent: Friday, June 15, 2007 1:46 PM > > To: Oracle-l > > Subject: Copying Large Amounts of Data From Prod > to Dev > > > > We are considering using EMC Symclone to copy data > from > > production to development databases. Does anyone > have any > > experience with this product used in this way? > Any 'gotchas' > > or words of wisdom? > > > > Pete Barnett > > Lead Database Administrator > > The Regence Group > > pnbarne@xxxxxxxxxxx > > > > > > > > > ______________________________________________________________ > > ______________________ > > Pinpoint customers who are looking for what you > sell. > > http://searchmarketing.yahoo.com/ > > -- > > //www.freelists.org/webpage/oracle-l > > > > > > > Pete Barnett Lead Database Administrator The Regence Group pnbarne@xxxxxxxxxxx ____________________________________________________________________________________ Be a better Heartthrob. Get better relationship answers from someone who knows. Yahoo! Answers - Check it out. http://answers.yahoo.com/dir/?link=list&sid=396545433 -- //www.freelists.org/webpage/oracle-l