Hi,
I have to extract data from a table based on the below sql
select * from xx where col1 = #var1#
The #var1# is a variable to be populated at run time (could be from a seq file or another table)
In server we can write 2 job first to have #var1# to be populated into a hashfile , then do a utilityhashlookup in the second job to use the variable to be put in the where condition.
How to achieve the same in parallel?? fileset Lookup stage will be inefficient
as as I need to select only few records out of millions.
Is there something similar to utility hash lookuo or pass data from one job to another??
I believe using a sparse lookup would work. You just have to select the value you want from the initial table, then lookup to the table you mentioned. select the lookup option to sparse in the DB connection and in the where clause do where col1 = orchestrate.source_column
Couple of ways to do it, you can use a sparse lookup as referred to in the previous postings. Or you can write two jobs, and put them both in a sequence. In the sequence you can cat the file and then pass the return value of the output as a parameter into the second job. If you are going to have more than one value that you need to have in the where clause you would need to go with the sparse lookup
Why move to a parallel job if you aren't really gaining anything by moving? Implementing the overhead of parallel processing for a few rows negates the performance improvement running in parallel can give.