We have some data import operations that originally dropped the indexes for efficiency, but now perform better with them left on during the import because we have transferred so much data now.
That's interesting Pete. My general experience has been, the larger the amount of data, the greater the need for dropping the target indexes. I guess one can't make such a blanket statement, and that there are a lot more factors involved in the copying speed than I know of. So it looks like one has to try different methods to see what works best. Thanks for the knowledge.