How to avoid full tablescan with max() function and group by

Question

Hi Exports,
This is my query:
Select max(dID) as dID
from stellent.Revisions
Group by dDocName
Plan
SELECT STATEMENT CHOOSECost: 3,154 Bytes: 2,336,150 Cardinality: 93,446           
     2 SORT GROUP BY Cost: 3,154 Bytes: 2,336,150 Cardinality: 93,446      
          1 TABLE ACCESS FULL STELLENT.REVISIONS Cost: 632 Bytes: 5,501,700 Cardinality: 220,068
How can I tune the query ?
Thanks a lot!

Answer 1

So I'm assuming it's slow and that's why you're asking for help and not just because it says full table scan. The problem is that you are reading every single row in the table regardless, so the best option is full tablescan. If you don't want to read every row in the table then I would suggest starting with a where clause.