vamsi

• Oct 25th, 2005

 

Merge is a stage that is available in both parallel and server jobs.

The merge stage is used to join two tables(server/parallel) or two tables/datasets(parallel). Merge requires that the master table/dataset and the update table/dataset to be sorted. Merge is performed on a key field, and the key field is mandatory in the master and update dataset/table.

jagadish

• Nov 3rd, 2005

 

  hi

i dont think so.

actually the merge stage in parallel job mainly used to merge the two or more data sets.

it will take one master ds file and n number of update ds files.

the output will one one final ds file +number of reject ds files as there r update files

maily for join

sivarao

 Profile Answers by sivarao Questions by sivarao

• Dec 28th, 2005

 

Merge stage is used to merge two flat files in server jobs.

Sharanu Patil

• Feb 28th, 2006

 

Merge is maily used to join two flat or sequential files in server jobs

shameerok

• Mar 22nd, 2006

 

Merge stage is a processing stage, it can have any no of input link and only one output link. It is having master data set and one or more data sets. The out put of the Merge stage is master dataset plus additional column from each update link. Where as Merge in Data stage server you can merge two flat file by specifying their location and name. the output will be join of two files. It is also like merge stage but only difference is that how the memory they use.