Archives of the TeradataForum
Message Posted: Wed, 31 Dec 2003 @ 14:10:01 GMT
Even if you use a batch size of one, each AMP would be assigned a single ID and there would be no sequential guarantee in the data load as AMP0 would be assigned ID 1 , AMP1 would be assigned ID 2 and so on (Note exact sequencing may be different). Sequencing is strictly dependent on sequence of hash values of each row being inserted.
Using Narayan's example:
Batch size =1
Amp0 : StartVal: 1 Last val: 1 Amp1 : StartVal = 2 Last Val = 2
|Copyright 2016 - All Rights Reserved|
|Last Modified: 28 Jun 2020|