![](//teradataforum.com/imgs/ptdfw1.gif) |
![](//teradataforum.com/imgs/ptdfw2.gif) |
Archives of the TeradataForum
Message Posted: Thu, 27 Sep 2001 @ 19:57:09 GMT
Subj: | | Re: Multiload performance |
|
From: | | David Wellman |
Ruth,
One option which may help you is to only drop/create some of the nusi's. The nusi too drop is the one which is most non-unique. This is
the nusi that will hit multiload performance the worst, so dropping/creating this one will give you the most "bang for the buck". I've seen
this approach taken a couple of times and gets the best of both worlds. BTW is there any mileage in upping the priority of the multiload
jobs?
Is using tpump an option for you? This won't load the data as quick as Multiload, but tpump gets affected by nusi's a lot less and may
allow you to load data during the online day - it has less impact on other processing. If doing this you can also use priority scheduler to
regulate impact of the tpump jobs.
Cheers,
Dave
| |