I want to read data from a join of multiple tables in salesforce.com. When I build the data flow using a Query transform , the joins are not pushed to the source datastore( salesforce.com). data services reads all rows in each table individually, then performs the join in memory on data services server side.
This causes performance degradation as some tables used in the join have millions of rows. So even for a final output of 10 rows, the job runs for 2+ hrs.
Using pushdown_sql () function in the where clause of the Query transform also does not work.
Using a SQL transform with the whole query does not work as at run time I get error "Unknown reader type 'Adapter'".
Is it possible to push down join criteria for salesforce.com tables. If so, how to do it. I am using 4.2 SP5.