yes william, that is what I mentioned on my 1st post. I think for the LIST partition oracle should do, but not sure for range and hash partitions.... -- thanks and regards ujang | oracle dba jakarta | http://ora62.wordpress.com On Thu, Nov 13, 2008 at 2:54 PM, William Robertson <william@xxxxxxxxxxxxxxxxxxxx> wrote: > I think Ujang's idea (correct me if I'm wrong) was that you could get the > distinct key values from the partition definitions stored in the dictionary. > However that information is not guaranteed to be accurate as you might have > empty partitions, or you might have previously used ALTER TABLE EXCHANGE > PARTITION WITHOUT VALIDATION. > > > -----Original message----- > From: Mark W. Farnham > Date: 12/11/08 16:49 > > Jared asks an interesting question, but I'm even more interested in why you > think > > > > "I think scanning Primary key > more efficient rather than roundtrip read PVC_U1 index and then again > read PVC_PK index, due to IOT mapping..." ? > > > > Your query only needs the seq column. All the column seq values are in > pvc_u1, which is smaller than the PK. > > So of course the CBO uses the smaller index. Since both SEQ and ES are in > the PK, I can absolutely guarantee that pvc_u1 is smaller than the PK, which > contains them and additional columns. > > > -- //www.freelists.org/webpage/oracle-l