Funnel Operation

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
amar.v023
Participant
Posts: 17
Joined: Thu Nov 19, 2009 12:55 am
Location: Hyderabad

Funnel Operation

Post by amar.v023 »

Hi All,

I have a requirement like,

haveing three operations 10,20,30(3 inputs) i want funnel and store into target.
o/p is
row1: 10,material number,-,-,-..........
row2: 20,material number,-,-,-............

and 30 operation will include when certain Material Number will match and o/p will be otherwise o/p will like above(only 2 operations):

row1: 10,material number,-,-,-..........
row2: 20,material number,-,-,-..........
row3: 30,material Number,-,-,-.........

when trying it's adding sequentialy and 30 operation is mixed with other material Numbers,

Any one have any idea pelase share.

DS is 8.1, Unix.
Amar
harishkumar.upadrasta
Participant
Posts: 18
Joined: Tue Dec 25, 2012 10:39 pm
Location: Detroit,MI

Post by harishkumar.upadrasta »

Please try a Remove duplicates after the Funnel stage based on the key colums to remove the duplicates.
Harish
vmcburney
Participant
Posts: 3593
Joined: Thu Jan 23, 2003 5:25 pm
Location: Australia, Melbourne
Contact:

Post by vmcburney »

Two things are happening in this job that is separating your Material Number row groups. First is the partitioning, default round robin partitioning is sending rows down different partitions so your 10, 20 and 30 rows end up on different partitions. Second is the Funnel is merging partitions back together again in an unpredictable order.

You could partition the job on Material Number - this might help keep the rows together but I am not sure what the Funnel will do with the partition merge. You could add a sort stage at the end of the job on Material number.
Post Reply