Re: Raw size

Hannu Krosing <hannu@tm.ee>

From: Hannu Krosing <hannu@tm.ee>
To: Ioannis Theoharis <theohari@ics.forth.gr>
Cc: pgsql-hackers@postgresql.org
Date: 2005-03-14T07:01:09Z
Lists: pgsql-hackers
Ühel kenal päeval (neljapäev, 10. märts 2005, 20:07+0200), kirjutas
Ioannis Theoharis:
> 
> Hi,
> 
> i have a table:
> 
> create table triples(
> 	att0 varchar(1000),
> 	att1 int4,
> 	att2 varchar(20),
> 	att3 varchar(1000)
> )
> 
> My table has 990 raws.
> 
> The (possibly wrong) way, with wich i compute the size of the table is:

Varchar fields (actually most *char and text fields) use only actual
length bytes + some overhead for tuple header + page header, so unless
you fill all varchar(1000) fields with exactly 1000-byte strings, you
should use less than that.


> Is there any compression or what?

Compression is not used for tuples under 2k, so there _may_ be
coimpression depending on your exact data and TOAST settings.

> 
> ---------------------------(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
-- 
Hannu Krosing <hannu@tm.ee>