How I should implement indexing strategy this fact table ? Contain about 5 mlns of rows
Is worth to add bloom filter index here also ? If so, in which way ?
How I should implement indexing strategy this fact table ? Contain about 5 mlns of rows
Is worth to add bloom filter index here also ? If so, in which way ?
I personally would just start only with the ZOrder only on machine_id
and day_id
and test the performance. Given that you have only 5 million rows, then most probably will create a single file for it, so bloom filter won't make any sense.
P.S. Bloom filters are good when you have a lot of distinct values where you're searching by the equality.