Partition and Partition point parallel data processing and data redistribution

When should I use Pass Through then?

For a "ORA-01555 - Snapshot too old error" we were suggested by Informatica support to correct the ORA_HASH function used in our SELECT queries. We were using as ora_hash(rowid,1)=0, ora_hash(rowid,1)=1, ora_hash(rowid,1)=2, ora_hash(rowid,1)=3 for SELECT query on a partitioned table. We corrected them as ora_hash(rowid,3)=0, ora_hash(rowid,3)=1, ora_hash(rowid,3)=2, ora_hash(rowid,3)=3.
When we resumed the failed job, 'Copy to Destination' stage of archive job picked lesser records than what was expected. For one of the tables, instead of 4 parallel queries only one was found to be running on Source database.

What could be the cause? Does it require EDM restart after making changes like above ?

Sorry to say.. couldn't take your accent ... too irritating ...

Could you please make video for all the partitions with real time scenarios ? Thanks

Very clear explanation on partitions, with this video my past doubts all cleared ,thank you very much

Very Clearly Explained. We need some more videos for other partitioning techniques.Thanks

Thanks a lot. Explained nicely.

Clean and Clear explanation, thanks a lot

WHY ARE YOU FAKING THE ACCENT ? ITS VERY ANNOYING AND NOT CLEAR ...I COULDN'T WATCH THIS VIDEO BECAUSE OF THAT , JUST TYPING THIS COMMENT AND GONNA QUIT.

Hi Vivek,
I followed it and created 2 partiion points. My session is taking data from source and transfering to a target. No transformation involved, just one to one. But what I see is that the target table has double the data. That means it is populating the target twice with same data which I don't want. Could you suggest the reason for this and how to solve it ?

Thanks in advance.
Khyama

The video is informative but the accent is quite irritating .. 

very nice video...please give us some inputs on technical implementation...