Re: doubt

Ühel kenal päeval, K, 2007-07-11 kell 19:08, kirjutas Greg Smith:
> On Wed, 11 Jul 2007, Narasimha Rao P.A wrote:
> 
> > Does postgreSQL support distributive query processing
> 
> Not internally.  It's possible in some situations to split queries up 
> across multiple nodes using add-on software.  pgpool-II, available at 
> http://pgfoundry.org/projects/pgpool/ provides an implementation of 
> distributed queries if your table has a type of key such that you split 
> across it, but it's relatively immature software 

Actually it is not "immature" at all, it is used 24/7 doing tens of
thousands of queries per second :P

But it is limited (by design) in what it can do - it is meant to run a
postgresql _function_ on one or more db hosts based on parameter
hash(es).

This can be used as a tool to construct a system which does distributed
queries, and also to distribute load on small OLTP queries over a set of
databases.

> and you would have to 
> look at it very carefully to see if that parallel query implementation 
> could fit your needs.

It has no parallel query implementation by itself (other than a special
running the same SQL on a set of hosts and merging the results), but you
definitely can progrem on using pgpool.

If you need something to distribute queries over a number of hosts
automatically, there is a commercial data warehousing product (based on
postgresql) available from greenplum, which does exactly this:

http://www.greenplum.com/index.php?page=greenplum-database


> --
> * Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD
> 
> ---------------------------(end of broadcast)---------------------------
> TIP 6: explain analyze is your friend