DSXchange

no because the source data is not partitioned at all. Haven't really tried it, why don't to test it and let us know the outcome.

vamsi.4a6 wrote:can i use same partition in transformer?If not what is the reason?

You could use same partitioning to force the downstream operator to run sequentially but why you want to do it that way, why not set the transformer stage to run in sequential mode? at the end you will end up with the same thing with either approaches.

with Ray's suggestion you will have an added advantage of "user friendliness", afterall you wouln't expect a person to understand that the tranformer stage is running on sequential mode when the job design show a partitioner on canvas but is removed out when the job runs due to same paritioning

Thanks all for your valuble suggestions and comments.Due to maintenance activity i did not try the job yesterday.Today i
tried and observed following points

sequentialfile--->sortstage--->transformer----->peek---->sequentialfile


1)I am getting correct output in both the cases

I.e In transformer with same partition and also running the Transformer in sequential mode

2)when i used Transformer with same paritititon , peek stage with auto partition
with All partitios=True. in the log all the input records are shown in partition 0 as below

Peek_26,0: ran:1 id:101 col1:xxx col2:95
Peek_26,0: ran:2 id:105 col1:sss col2:92
Peek_26,0: ran:3 id:103 col1:vvv col2:91
Peek_26,0: ran:4 id:102 col1:yyy col2:80
Peek_26,0: ran:5 id:107 col1:mmm col2:51
Peek_26,0: ran:6 id:106 col1:eee col2:50
Peek_26,0: ran:7 id:104 col1:ddd col2:40
Peek_26,0: ran:8 id:109 col1:asdf col2:2
Peek_26,0: ran:9 id:108 col1:adsa col2:1

Log is not showing any information realted to partition 1,2,3 and i am using 4 node configuration file


Once again Thanks all for your valuble suggestions since i am beginner for Datastage