Can anyone explain how Attunity Connect returns rows from a query against RMS and RDB datasources? Does it first collect all rows to memory? spill over to disk? and then return the data? Can it stream for simple queries such as a basic "select * from table." We're trying to ETL (via Informatica PE) all rows from a RDB datasource and it hangs and we are observing a large hash overflow file getting created. Should we paginate the query? maybe every 1000 rows and walk down the table that way? It's also possible the hash file is getting created from a different query against RMS datasources with a join. Can't really tell why/where this file is growing so large. What's the story behind the hash/overflow.
↧