Hello,
Considering the CLUSTER operation on a frequently updated table, if I
have 2 indexes on the table how do I choose one of them ? or is it
possible to have CLUSTER take both into consideration...
my table is read from based on two columns: a 'timestamp' integer column
(actually a UTC integer timestamp) and a text columun called 'var' that
separates raw data from each other... something like a 'product type' in
a wharehouse example.
All queries that read from the table specify a time range and also the
'product type'. I have indexes on both.
Evidently no CLUSTERing should be worse than clustering on any of them,
but is there a way to have the two or choose the most profitable ?
thanks
Joao