Archives of the TeradataForum
Message Posted: Sun, 02 Mar 2008 @ 09:21:58 GMT
Hello Dave, Michael, Victor
Thank you all for your replies.
I tried the experiment in the DEV env by setting the table to NUPI and MULTISET. The time difference was just a few minutes and there was hardly anything running in the DEV environment. So I am guessing that if I habe 2-3 MLOAD jobs running the time would be the same. Also, we cannot have UV errors because the INFA mapping ensures the key generated is always unique. I just don't understand how this could be reduced. Does it make a difference to MLOAD if the target table size is huge (my guess is it should not)? Also, like I had said, the EVENT table has 20 fields out of which 12 are NULL always. Not sure if that is the issue.
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|