Answer / ankit kansal

Hi, if you know the number of duplicates coming your from source then it's easy to move the duplicates to the defined target using router transformation easily,
But if you do not know the number of duplicates then first you must sort the data on the value of duplicate column and then using Transaction Component as available in informatica you can create n no of targets depending upon the values encountered.

http://deepinopensource.blogspot.in/

Answer / kiran kumar

hi
put the hash partition in active stage then
make three node configuration then u got the output

Answer / suneelbabu.etl

can u mention some more input records as well output

Answer / sher

something like below should work...

Job#1
Read in transformer to add a new column sequence the value of sequence will be initially 1 and increment when key changes…
so output(A) will look like…
1 1
1 1
2 2
2 2
3 3
4 4
B 5
under job sequence, use a job activity to read(A) use tail -1 and | cut to read 2nd column. we get 5..
Now, use start loop activity and end look activity stage in job sequence to create a loop from 1 to jobactivity.output incrementing by 1..
The look will now run 5 times.. inside the loop, call a job and pass the value i.e 1,2,3,4,5 each time it runs.
Inside the called job, read the input file(A) pass to transformer and then use constraint to write only records where 2nd column = variable passed and drop column 2..
Output of transformer.
1st loop output
1
1
2nd loop output
2
2
and so on...
when writing the file, use the passed variable in file name, so file name is distinct

Answer / narayan

just add the mod function like this =MOD(inputcol|3) and then load into different targets

Answer / sivakeshava

hi friends by seeing the source we need 3 targets....which id duplicates 1,1 is one target, and 2,2 is 2nd target and 3 is another target
Seqfile------------ >copy -------------- lookup(innerjoin) ------------- filter

Aggregator---------------- > filter
(count) where clause

copy to aggregate then filter to lookup then will give 3 targets

Answer / ankur

Thats a good question... but unfortunately I dont know the ans.. Datastage gurus- can anyone help this guy and myself to know this question's ans.. Appreciate your feedback please!!

Answer / vaibhav

hello Ankit,

I need the ans in datastage..how this could be implemented in datastage?

Answer / vaibhav

Actually it is like:
I want as many target links as much u have types of duplicates in my input file.. Pls help guys..

can any one tell me how to install datastage 8.1 in windows xp with wizard

2 Answers  

WHAT ARE PERFORMANACE SETTINGS YOU HAVE IMPLEMENTES IN YOUR PROJECT?

2 Answers   GE,

What is usage analysis in datastage?

0 Answers  

In work load management there are three options of Low priority, Medium priority and High Priority Jobs which can be used for resource management. why this feature is developed when there is already jobs prescheduled by scheduler or autosys. what will be the use of workload management then?

0 Answers  

how do u reduce warnings

1 Answers   CGI,

How to add zero "0" before record in a field?

4 Answers  

What could be a data source system?

0 Answers  

Hi guys, Design job sequence, we have 3 sources, in that 1st source in abort then only run the remaining sources.. How please design the job. Thanks.

2 Answers   IBM,

What is difference between join, merge and lookup stage?

0 Answers  

What are the difference types of stages?

0 Answers  

AGGREGATOR default datatype

0 Answers   CTS,

which unix commands mostly used in datastage

3 Answers   HSBC,