On Aug 12, 2010, at 15:56 PM, Paul Ramsey wrote:

> Because of transaction handling, updating every row in a database is
> the equivalent of a full load. It might actually take longer than the
> initial load.
> 
> I'd suggest something like
> 
>  create newtable as select st_makepoint(x,y) as point, a1, a2, a2...
> from oldtable
>  drop oldtable
>  alter newtable rename to oldtable
> 
> P.
> 


Sorry Paul but I got very surprised with you answer. I can't understand why to 
create all the records again is faster than update them. 

Can you please give some links to read about it?

We are getting performance problems with some table updates and I guess that 
could be the reason.

Thanks in advance.

--
Mauricio Miranda
Software Engineering Manager
http://www.xoomcode.com
_______________________________________________
postgis-users mailing list
[email protected]
http://postgis.refractions.net/mailman/listinfo/postgis-users

Reply via email to