one of the first things to look for in database performance tuning
is the existence (or nonexistence) of indexes (for SELECTs, that is). 
 they are perhaps the "#1 most used" optimization for relational 

if you can, i would suggest first tracking down the query text itself.  
sybase and oracle have mechanisms for logging full query text (oracle 
to a file, sybase to the sybsecurity database in sysaudits_0x tables),
mysql and postgresql most likely have the same features.  if not, it
would be a trivial task to have query text logged to a file.  next,
i would determine if there are indeed indexes on the tables in the
query.  if there are none, try adding some. if there are some, then
try adding more. :)  (without getting into too much detail, try 
adding indexes on columns that are commonly joined on or restricted
on...primary keys for tables is also nice).


On Thu, Sep 18, 2003 at 12:08:51PM +0100, Supote Leelasupphakorn wrote:
> To all,
>    As a newly DBA, I really don't know how I deal with
> this problem. My problem is not so long ago, my database
> server seem to overloaded. It take me a time to find
> the cause of problem. I realize that some program don't
> queried wiht inappropriated SQL statement. I mean they're
> not efficient one.
> AS DBA how do you solved this problem?
> Thanks in advance,
> ________________________________________________________________________
> Want to chat instantly with your online friends?  Get the FREE Yahoo!
> Messenger
> _______________________________________________
> [EMAIL PROTECTED] mailing list
> To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Attachment: pgp00000.pgp
Description: PGP signature

Reply via email to