Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Wed, 26 May 2004 @ 18:31:36 GMT


     
  <Prev Next>   <<First <Prev
Next>
Last>>
 


Subj:   Re: Identity Columns
 
From:   Dieter Noeth

Hi Joe,

if you look at a single AMP it's behaviour is similar to e.g. Oracle's implementation: Retrieve a bunch of numbers and then use it. If there's a Rollback or system restart those values will be lost.

But Teradata is a parallel DBMS, so each AMP with rows from the source table requested a bunch of numbers from dbc.idcol (by default it's 100000 values and can only be modified globally, not per table) and assigned it to those rows. Your result indicates that you run that query on a 12-AMP system.

Now for the confusing part: do a INSERT ... VALUES and the next number will be 1200001 ;-)

     single row inserts -> PE assigns identity values
     insert/selects -> source AMP assign identity values

Dieter



     
  <Prev Next>   <<First <Prev
Next>
Last>>
 
 
 
 
 
 
 
 
 
  
  Top Home Privacy Feedback  
 
 
Copyright for the TeradataForum (TDATA-L), Manta BlueSky    
Copyright 2016 - All Rights Reserved    
Last Modified: 15 Jun 2023