Restrict data to 1000 rows

[system]

How about the idea in this post:-

120×120 10.4 KB

I need to take a random sampling of records from a very large table. I would like the work to be pushed down to Oracle. The best way I know involves writing a subquery such as: select pk_id from (select pk_id from table order by dbms_random.value()) where rownum <= 10 What’s the best way to accomplish the same thing in DI without sending all of the records to job server for the sampling? Thanks! tangel (BOB member since 2008-05-12)

---

Nemesis (BOB member since 2004-06-09)