I have a table with a date field that lends itself to partitioning. However this is not the field that is used in most queries, so I currently have 2 indexes on other columns in the table. I want to partition the table mainly to speed up deletes, which are done based on values in date field. This allow me to drop a partition rather do a normal delete. Does it make sense to 1. Partition the table by the date column 2. Create a local partitioned index using that column 3. Create 2 other indexes on the 2 columns used for select statements. Should these be non-partitioned or global partitioned? John -- //www.freelists.org/webpage/oracle-l