At 10:00 PM 3/17/2009, Harald Armin Massa wrote:
Merlin,
> I agree though
> that a single table approach is best unless 1) the table has to scale
> to really, really large sizes or 2) there is a lot of churn on the
> data (lots of bulk inserts and deletes).
while agreeing, an additional question: could you please pronounce
"really, really large" in other units, like Gigabytes or Number of
rows (with average rowlength in bytes, of course)
That is: what table size would you or anybody consider really, really
large actually?
Tiny: fits in CPU cache
Small: fits in RAM
Big: multiples of RAM.
Large: (size / storage bandwidth ) is measured in minutes.
Huge: (size / storage bandwidth ) is measured in hours.
Humungous: (size / storage bandwidth ) in days or larger units.
That said, the active working set might be a lot smaller than the
table, in which case you might prefer to use the size of the working
set (except when you are doing stuff like full backups or restores).
Link.
-
Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
