Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Thu, 06 Mar 2008 @ 15:08:26 GMT


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


Subj:   Re: No Stats better then Current Stats?
 
From:   Victor Sokovin

  Most of us are not in denial at all. I have seen this case, and other oddities, like having an index with stats sometimes creates a better plan, than just stats on the column...even though it does not use the index :)  


Aha, that is certainly reassuring, especially if somebody confesses in public the sighting of the latter kind. I once confessed it (years ago) and got spooled away with both theoretical and practical considerations explaining why exactly I could not have seen anything of that kind. But I never gave up to find more members of this club as I thought I understood the background of the sightings. It was too difficult to explain it (although I tried), especially not having access to the original code, but it was a good test question to see who was a member of the club and who was repeating somebody else's opinions (there are other categories, of course, as well).


  Bottom line is, it is always suggested (strongly) that you need to evaluate any changes to ensure that they result in the desired effect.  


That is certainly true; prioritizing is the key and evaluation of the side effects is not always easy.


  So we are with you; we want the best optimized plans also. After all, we have sla's to meet too.  


That's again very reassuring; thank you for the confirmation. The competition is there, and it is getting tougher and tougher, it really is. Keeping Teradata going is probably not as easy as it used to be.

Speaking of more general future plans it would be good to see you guys changing your business model as well (a modest wish this one is, yes?). In my opinion, Teradata product line should be extended to cater for smaller businesses, and the pricing should not be such a prohibitive linear function of the storage as it is now. You should dig into a broader market and lose dependency on a few large clients (speaking of priorities). Simply because it is a more interesting market which would provide you more feedback and challenge you with more tasks. It should be more fun for you and eventually more fun for the Teradata users in general. Otherwise, the base is getting too small. Just a thought as a follow-up on the stats question.


Victor



     
  <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