Simple test was done using sqlldr: Load a table in direct mode using rows=3D1. While the load is running and saving each row, non of the saved data is = available to the other sessions. Aborting the sqlldr session made the saved data available to other = sessions. Regards, Waleed -----Original Message----- From: jaromir nemec [mailto:jaromir@xxxxxxxxxxxx] Sent: Wednesday, April 28, 2004 4:57 PM To: oracle-l@xxxxxxxxxxxxx Subject: Re: Informatica Bulk Mode behavior H, >There might be the need for partial data availability during the bulk >load to achieve some multi-tasking (run other processes on the saved >data), in this case some warning about the effect of <low commit >interval + Oracle> would be useful in the documentation. I guess this is not only problem of parallel processing. What would = happen, if the Informatica session crashes after the x-th commit point in case = that "OCIDirPathDataSave()" is used? Can the Informatica session be restarted from the last commit point? I shortly consult OCIPG, but I don't found a definitive answer. regards Jaromir D.B. Nemec PS: In our shop we have a convention to use "very high" commit interval = in this case. ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //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@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------