Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Mailing Lists -> Oracle-L -> RE: long runtimes cured by cutting hash_area_size by 90% - ???
Probably when the hash area size is big, this encourages the optimizer to do
a hash join or may be many but on the other hand when it's small the
optimizer chooses to avoid an expensive hash join by probably a nested-loop.
You need to get the execution plan twice: one with the big hash area size and the other with the small one.
Regards,
Waleed
-----Original Message-----
Sent: Friday, April 11, 2003 5:44 PM
To: Multiple recipients of list ORACLE-L
Attached are 3 PLANs output by tkprof after SQL_TRACE. This is Oracle 8.1.7.4 on Sun Solaris, NO parallel query.
Attached also are
Queries are generated by a tool, Cognos.
Note: table RI_XLR_T has 1.8M rows, others are ALL under 60k.
tkprof PLANS -
13 secs., fastest w/ 1Meg HASH_AREA_SIZE.
(Similar times seen for 2Meg, 500k)
6 mins. 35 secs., slowest w/ 100Meg HASH_AREA_SIZE. There's a killer HASH JOIN at the innermost point of the PLAN .. (Similar times seen for 8Meg, 10Meg, 20Meg)
Hash_Area_Size runtime (faster when smaller):
> 10,000,000 6:16 ( 6+ mins )
> 1,000,000 0:15 ( 16 secs)
This does NOT use parallel query.
It is repeatable.
Table Rowcounts:
xlr: 1.8M xcp: 61k mcp: 44k xru: 30k xrx: 30k
In advance thanks if you can shed light on the mysteries of CBO, etc. etc.
-- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Khedr, Waleed INET: Waleed.Khedr_at_FMR.COM Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services --------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).Received on Mon Apr 14 2003 - 09:49:14 CDT