Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: Informatica Bulk Mode behavior

RE: Informatica Bulk Mode behavior

From: Thomas Jeff <jeff.thomas_at_thomson.net>
Date: Thu, 22 Apr 2004 11:38:34 -0500
Message-ID: <358728A276824E419580403633AABFD0021F614A@INDYSMAIL03.am.thmulti.com>


Daniel,

With all due respect, why would this make a difference?? The tablespace is locally managed,
128 MB uniform extents, 4GB in size. We know from the normal mode process that the data=20
uses around 1.5GB. =20

The table itself has about 40 NUMBER columns used to store various quanity info.

Jeff

-----Original Message-----

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Daniel Harron Sent: Thursday, April 22, 2004 10:44 AM
To: oracle-l_at_freelists.org
Subject: RE: Informatica Bulk Mode behavior

Can you provide the DDL for the table and tablespace?=20

-Daniel

--=20

Daniel Harron
Database Management
IPsoft, Inc.
daniel.harron_at_ip-soft.net
http://www.ip-soft.net/
Phone: 888.IPSOFT8
Fax: 801.681.7664
-----Original Message-----

From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org]
On Behalf Of Thomas Jeff
Sent: Thursday, April 22, 2004 10:33 AM
To: oracle-l_at_freelists.org
Subject: Informatica Bulk Mode behavior

List,
Is anyone familiar with how Informatica's bulk mode works? =20

We have a situation where I created a 4GB tablespace for a new and simple truncate/load operation from Informatica, around 7 million rows estimated to take up 1.5GB. =20
 =20
Using bulk mode, which appears to be a direct load (in the SQL cache, the INSERT statement has a hint that I've never seen before: SYS_DL_CURSOR which I assume stands for Direct Load), they run out of space in the tablespace after about 200K rows have been inserted.

If I then manually rebuild the table, the 200K rows gets compressed back down to one extent.

So there's a lot of either empty or preallocated space. Thinking somehow
the high-water mark was the culprit, I manually truncated the table before=20
they reran their job. =20

Still the same problem. =20

If the job runs in 'normal' mode, which is row-by-row processing, it runs fine, although of course, performance is quite poor.

Any ideas/experience as to how this bulk mode operates?

Thanks.



Jeffery D Thomas
DBA
Thomson Information Services
Thomson, Inc.

Email: jeff.thomas_at_thomson.net

Indy DBA Master Documentation available at: http://gkmqp.tce.com/tis_dba




Please see the official ORACLE-L FAQ: http://www.orafaq.com

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.
--

Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html


Please see the official ORACLE-L FAQ: http://www.orafaq.com

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.
--

Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html

-----------------------------------------------------------------
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request_at_freelists.org
put 'unsubscribe' in the subject line.
--

Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
Received on Thu Apr 22 2004 - 12:08:57 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US