Excuse my ignorance on this (there is a reason why I still have "1" next to my DS profile :D ) by "dumping" do you mean just doing a link from hash file to sequencial file? Of course, I am referring to the dumping that doesn't involve writing the BASIC subroutime you mentioned.kcbland wrote: Full scanning and dumping a hash file to a sequential file is VERY FAST. DataStage adds overhead, he could consider writing a very simple BASIC subroutine to dump a hash file to a delimited text file. You don't have to worry about formatting, just do delimited. Simply select all rows from the hash file, then read each row, convert @AM to "|", and WRITESEQ out the row. Think of it as a mindless bulk-unload.
Thanks..