I've got this issue with a query in SSIS. From a table in SQL Server I'm getting over 25000 different identifiers. These identifier are associated to many values in a table in one Oracle Database. This is the schema that I have implemented for doing this.
The problem is that some days the identifiers can be over 45000, and at this point perform a loop for every one is not the best solution (It can take to much time to get the result). Previously I have performed another query where from the SQL statement, I am creating and sending a unique row with all the values concatenated and then I have recover this unique string from an object and use it to create the query in the ODBC Source that invoke the table in Oracle: something like this:
'Select * from Oracle_table' + @string_values
with @string_values = 'where value in (........)'. It works good because the number of values is small enough to be used, like 250. But in this case I can not use this approach because the number is really big and obviously the DBA of Oracle is going to cancel the query.
So I wonder, how can I iterate over the object getting only a few number of values everytime, something like 300 or maximun 500, to avoid the cancelation of the query but at the same time doing the minimun number of loops. Or if there another solution to this question I would appreciate your help.