A table with 4000 partitions

What are possible advantages/disadvantages of using very large number of partitions, let's say, 4000 or more, in DSS environment? I see following
Advantages:
1. Reduces the need for indexes, and potentially no indexes at all 2. More flexibility in removing information by using "alter table truncate partition"
3. It is possible to run several direct loads in parallel in differet partitions
4. More flexibility with tablespaces/datafiles

Disadvantages:

1. Bigger shared pool
2. Higher ENQUEUE_RESOURCES
3. Some operations are slower, for example, snapshot refresh

Anything else?

Again, I am talking about DSS environment, with big "historic facts" table
being gradually populated from one end (let's say by direct loads) and truncated at the other end, the information is never updated, users only access this table by running reports, so there is never a need
to return just a few rows.

I saw a note on the Metalink that says that the maximum number of partitions
per table is 64,000.

Thanks,
Sev Received on Mon Jan 28 2002 - 17:05:50 CST