Hash file equavalant in parallel Job

[edward_m]

I was trying to design a job to load data into oracle table, before table load i would like to compare the records based on the primary key which consists of 4 keys, if record already exists then log those records into file, if not insert those records.
In server job i would have create hash file based on primary key then in transformer use hash lookup and constraint.
Is there any similar functionality in parallel stage, i think we could use lookup stage but not sure which constraints to use.
could you please point me in right direction?

Thanks in advance for your help.

[ray.wurlod]

A hash file is a file used for shaping a block of hash.

DataStage has hashed files.

There is no equivalent in parallel engine.

But it seems to me that a Lookup stage with a reject output link would serve your purpose adequately. Just be aware that the reference data set must fit into memory.

[edward_m]

Thanks for your help.