Thread: shared temp tables

shared temp tables

From
mlw
Date:
I love the fact that temp tables do not exist in every PostgreSQL session,
don't get me wrong. 

The issue is this: most "web environments" have the idea of a session. A
session management scheme based on PostgreSQL exposes PostgreSQL's worst
behavior. Small amount of records, high update/delete rate for each record. So
much so, that it probably isn't realistic to replace something like Oracle with
PostgreSQL in this environment.

Do "temp tables" suffer the same delete/update behavior of marking the row as
deleted and adding another row? Thus requiring vacuum periodically. 

If not, should/could there be a way to create a temp table that is globally
visible?


Re: shared temp tables

From
Bruce Momjian
Date:
> I love the fact that temp tables do not exist in every PostgreSQL session,
> don't get me wrong. 
> 
> The issue is this: most "web environments" have the idea of a session. A
> session management scheme based on PostgreSQL exposes PostgreSQL's worst
> behavior. Small amount of records, high update/delete rate for each record. So
> much so, that it probably isn't realistic to replace something like Oracle with
> PostgreSQL in this environment.
> 
> Do "temp tables" suffer the same delete/update behavior of marking the row as
> deleted and adding another row? Thus requiring vacuum periodically. 
> 
> If not, should/could there be a way to create a temp table that is globally
> visible?

Temp table are the same as real tables have have the same update
behavior.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026