| Gather Statistics - too much time taking [message #203836] |
Thu, 16 November 2006 09:14  |
Brayan
Messages: 315
Registered: June 2002
|
Senior Member |
|
|
Hi,
I have gathering statistics from a table having 8000000 records. It is taking aroung 1.5 Hour to gather status. Is ther any way I can decrease the time for Gather stats..
I'm using
DBMS_GATHER.GATHER_TABLE_STATS('User1','Table1');
Thanks
Brian.
|
|
|
|
|
|
|
|
| Re: Gather Statistics - too much time taking [message #203979 is a reply to message #203836] |
Fri, 17 November 2006 03:29  |
Brayan
Messages: 315
Registered: June 2002
|
Senior Member |
|
|
Hi,
Is this the efficient way of analyzing
DBMS_STATS.GATHER_TABLE_STATS(ownname=>'user1', tabname=>'table1',degree=>16);
Here degree(degree of parallelism) is itself the parallel cluase or some other paralle clase exists.
Regards,
Ronald.
|
|
|
|
| Re: Gather Statistics - too much time taking [message #204036 is a reply to message #203836] |
Fri, 17 November 2006 07:42 |
Brayan
Messages: 315
Registered: June 2002
|
Senior Member |
|
|
Hi,
In 8 CPU db server I have analysed the table having 20000000 records using following...
dbms_stats.GATHER_TABLE_STATS(ownname=>'TARGETDB',tabname=>'DEPOSITS', degree=>16);
It is taking around 4 hrs. I do not have this much big window to analyze. Even I have many tables with this size. Please suggest what I should do to reduce time.
Brian.
|
|
|
|
|
|
] 
Blog
Search
Help
Members
Register
Login
Home
