Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Mon, 05 Nov 2007 @ 15:39:34 GMT


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


Subj:   Mload: Taking very long time to run
 
From:   Anomy Anom

<-- Anonymously Posted: Mon, 5 Nov 2007 10:36 -->

Hi All,

I am inserting and deletiing certain rows in a table using Multload. Since the table data is highly skewed the Mload delete step is taking more than 9 hrs to complete.

My target table has non-unique pirmary index. It is not possible to change the PI of the table. Though I noticed that if I change the PI from 2 to 3 columns the table becomes failry distributed on all the amps using the query

     *Select * hashamp(hashbucket(hashrow( NAME, ORIGIN, HO_NO))) ,*Count*(*)
     *From   * MCKSG03_RESP
     *Group   By* 1;

So if change the PI from (NAME, ORIGIN) to (NAME, ORIGIN, HO_NO) the distribution of the rows is fairly even.

But I cannot change the PI of the table.

If I define a non-unique secondary index on (NAME, ORIGIN, HO_NO) will my mload performance improve.

Is there any other way I can improve the performane of my Mload?



     
  <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: 27 Dec 2016