DSXchange

Hi

I have requirement where I load data into target
Without landing data into datasets I have to load to fact table
With 10 lookups and 10million records how can we achieve this

We don't have enough space on dataset directories

Thks

By creating job designs that don't load data to Data Sets?

Hi
From my understanding if you source from database and loading into target database in same job performance will be very slow
Do you have any sample design approach that can be followed

Thks

Your understanding is... incorrect. Or best case incomplete.

ETL from and to the same table can be slow (mainly due to locks or, worse, self-deadlocks).

ETL from and to the same database should not be a problem, if source and target are different tables and the database server has sufficient capacity for the number of connections being requested.

if you need to, can you load a temporary table in your target or source or other DB?

You can run straight through, of course, if there isnt anything in your job that runs out of resources. If there is, you may have to find a workaround (like the temp table).

You can also change your extract to get the data in pieces, and run the job a few times with different where clauses.