Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Thu, 03 Aug 2006 @ 14:17:55 GMT


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


Subj:   Teradata backup and stats performance problems
 
From:   Madagani_Rao

Hi All,

I am new at Teradata and looking for help in the form of a suggestion to an issues explained below.


Back-up:

Currently backup for our biggest table O_PSTEB_BAL takes 6 hours 15 minutes to complete (we are running "all amp backup"; therefore, backup for single table cannot run across multiple tape drives).

If we combine "cold data" and "hot data" into single table backup will run for 7 hours 30 minutes. Every consecutive month backup will take +/- 20 minutes more to complete due to table growth.


Stats:

Currently we are running stats on "need to have basis" when performance problem occur. Requirement is that we run regular daily/weekly stats for all tables and at list for following columns:

primary index columns

partitioning key columns

all known columns used in joins

combinations of columns known to be frequently joined

any column which features in WHERE conditions


Running this type of stats for our big tables is very long process and cannot be scheduled to run regularly after mload.

Proposed solution:

Split our big tables into "hot tables" and multiple "cold tables" after Teradata upgrade is completed.


Benefits:

Backup for big tables will be spread across multiple tape drives and will complete in +/- 3 hours using 10 of 12 available tape drives.

Daily stats will run for "hot data" only and we will be able to schedule regular daily/weekly stats for all "hot tables".

Only backups of "hot tables" will be used for DRP and to restore/refresh future test/pre-prod system.

Any thoughts or suggestions ? Thanks in advance!!!


Regards,

Srinivas



     
  <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