On Sat, Mar 19, 2005 at 07:05:53PM -0500, Tom Lane wrote:
> Alvaro Herrera <[EMAIL PROTECTED]> writes:
> > We probably also need multi-table indexes.
> As Josh says, that seems antithetical to the main point of partitioning,
> which is to be able to rapidly remove (and add) partitions of a table.
> If you have to do index cleaning before you can drop a partition, what's
> the point of partitioning?

Why would you need to do index cleaning first? Presumably the code that
goes to check a heap tuple that an index pointed at to ensure that it
was visible in the current transaction would be able to recognize if the
partition that tuple was in had been removed, and just ignore that index
entry. Granted, you'd need to clean the index up at some point
(presumably via vacuum), but it doesn't need to occur at partition drop
Jim C. Nasby, Database Consultant               [EMAIL PROTECTED] 
Give your computer some brain candy! www.distributed.net Team #1828

Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"

---------------------------(end of broadcast)---------------------------
TIP 9: the planner will ignore your desire to choose an index scan if your
      joining column's datatypes do not match

Reply via email to