I have a dimension with 11 million records, and there are some facts in it,
so I designate it as the fact dimension since it will save storage and some
other benefit?

Now I am facing the issue of how to incrementally update the fact dimension
more efficiently. It seems every time I incrementally update the dimension
(even there are few or no new records inserted), it will take same or longer
time to process vs. initial full process. My question: is there any better
way to deal with this situation?

Since this is fact dimension, it is also a fact, on which I have partition
defined, so I wonder if I could just incrementally update the lastest
partition on the fact side, will this achieve the same purpose of incremental
update on dimension since they are actullay same. Or do I require to
incrementally update the dimension part?

Thanks,