Efficient delete from table with compound keys?

Using II 9.0.4 (hp2.us5/105)

What is the most efficient way to do the following? TableA is btree with col1, col2, col3 being the compound key and has about 12.5M rows of data. TableA is also compressed and stored across 3 locations if it makes a difference.

delete from tableA where col1+col2+col3 in
(select col1+col2+col3 from tableB)

Thanks

Dennis
d underscore roesler at agilent dot com
.

Relevant Pages