query optimization

I have one table (A) with ~30,000 rows in it and a second table (B) with ~85,000 rows in it. Each table contains an ID column, and each table in indexed on that ID column (these are not primary key type ID's; there are duplicates, so the indexes are not unique). I want to delete from table A all rows where the ID does not exist in table B. When I'm done, I've deleted ~22,000 rows (>70% of the rows) from table A. The problem is the query takes most of the day to run, so I'd like to optimize it. Because this is a batch deletion I want best throughput rather than best response time. As far as I know we don't generate analysis statistics, so I believe this means we're using the rule-based optimizer. Would a hint help me?

TIA,
Karen Kluge
UC Davis
Davis, California Received on Wed Aug 13 1997 - 00:00:00 CDT