Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Mon, 28 Jul 2003 @ 12:17:59 GMT


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


Subj:   Re: The size of Volatile Tables
 
From:   David Wellman

Craig,

I doubt that dealing with a Volatile table makes any difference. I'd suggest that (for diagnososis purposes) you treat this much like any table.

So...

-- if this is an Insert/Select request, do you know if it's the Insert or the Select phase which took the time?

Assuming it's the Insert

-- check the primary index of your table. Is your data skewed? How many rows (Max and Average) per PI data value?

-- do you have any secondary indexes on the table? Are they highly non-unique?

-- was the table empty prior to populating it? If it wasn't and it's a requirement to add to a populated table, you could use the "NO LOG" option whic prevents any transient journalling entries being written (BUT you have to sort out the data after any failures).

If it's the Select,

-- check which stage of the explain plan takes the time.

-- try indexes, statistics etc etc


Cheers,

Dave



     
  <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