Based on your experience with Oracle, what would be the best type and index options you would set in a DATE type column?
- I don't have to look for a partitioned index.
- This is the log for the table.
- You really do not care about the unique identifier as the primary key (in fact, the date is close enough to be unique in most cases, but due to its nature it will never be).
Will the creation of a cluster index be fair?
I am interested in optimizing query execution such as SELECT * FROM Log WHERE [Date]> '20 -06-2009 'ORDER BY [Date] DESC without slowing down the insertions in bulk. (by the way, in the real world, I would use the correct TO_DATE syntax to avoid truncation and lack of index)
Greetings
source
share