Archives of the TeradataForum
Message Posted: Fri, 06 Jun 2003 @ 10:56:19 GMT
Your assumption is correct. Specifying the INDEXES keyword will mean that the archive takes longer (simply because arc has more work to do).
My view on this subject has always been that you archive more often than you restore, so I typically do not use the INDEXES keyword thereby saving time on the most frequently repeated operation. I then accept the longer time it takes to do the restore, but (hopefully) that doesn't happen too often.
Given this scenario (i.e. NOT specifying the indexes keyword), my normal approach to the restore process is to use the NO BUILD option on the RESTORE and then run the BUILD command after the restore to build the secondary indexes. It's a little while since I delved into this area in detail, but I've seen this work very well (and quickly) especially when the archive file contains multiple tables with SI's. I think you'll find that if the archive file contains, for example, a single table with SI's then there will not be much difference between building the si's as the table is restored or specifying the NO BUILD and then running a separate BUILD command.
Hope that helps.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|