Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Wed, 04 Nov 2009 @ 10:24:03 GMT


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


Subj:   Going by book or be pragmatic: Collect Stats
 
From:   Anomy Anom

<-- Anonymously Posted: Wednesday, November 04, 2009 02:24 -->

Hi Gurus,

There are tables which are having almost millions rows, by which I mean tables are fairly big, though not large. There are queries on such tables which are somewhat complex, mostly INNER JOINs. Except the PI columns there are no stats on these tables still the response time of these queries is very good, less than a second generally.

In this scenario when I run explain plan using (diagnostic helpstats), it gives a list of columns/combinations for which stats should be collected.

If these stats are collected there might be slight improvement which would not be noticeable as already queries, as I said earlier, are taking less than a second.

The tables are part of staging database where no history is being maintained so the data volume may increase or decrease a little but not exponentially as may happen in history tables.

In such a scenario should I collect all these statistics, and burden system with more work for regular stats update or let system run as it is as the response time is already good and wait for time when system starts behaving badly and then collect all those stats. Or I should collect all the stats required even though it may not improve performance much now but keep my system well theoretically.

I want experts advice on this dilemma of mine. Going by book or be pragmatic.


Regards



     
  <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