DSXchange

Hi All,
I am using Parallel Extender 7. I am extracting data from DB2 and the number of records is 5500000000 (550M). I am putting this data into DB2. Now my job aborts at 250M (250000000) the record. Now when doing a reload I want to start it from 2500000001 th record. Kindly let me know how to achieve this.

You could use a constraint to start at @INPUTROW > StartRow where StartRow is a parameter. I am not sure if this works in PX.

I'd be more concerned with finding out why it aborts, and correcting that situation. How do you know that all of the 250M records were successfully loaded?
That said, the approach is basically to determine the row counts - there are many ways to do that - and to use the appropriate one as the start point for your next run.

kduke wrote:You could use a constraint to start at @INPUTROW > StartRow where StartRow is a parameter. I am not sure if this works in PX.

Thanks Duke.We'r trying the same.

Hi,

Do not use @INPUTROW in PX partitioned jobs because the value of @INPUTROW will be different in different partition. You can use Surrogate key Generator stage to create a runtime number. It is consistent across the partition.

Also you can control the commit level in DB2 stage.

Regards
Amitava