Archives of the TeradataForum
Message Posted: Thu, 19 Jul 2007 @ 11:24:44 GMT
This is probably an option within an ETL tool such as Informatica or Datastage?
I checked the MultiLoad manual of Septembet 2006, and the option wasn't mentioned there.
If it is indeed the ETL tool option then you are right about its meaning.
If you use a pipe you also save space on your ETL sever as no intermediate file is created.
As to the performance, just try different options and see what happens in your particular case. You are supposed to be faster when you use pipes but there can be cases when the difference is not noticeable (the bottleneck is in the data export part of the process, not in the ML).
|Copyright 2016 - All Rights Reserved|
|Last Modified: 28 Jun 2020|