Re: size of bytea + performance issues

On Tue, Jan 31, 2006 at 09:15:18AM -0800, tschak wrote:
> Allright... do you mean an alignment with an offset of 4/8 bit or byte?
> If it is just bit I cannot really follow the calculation...
> Nevertheless it sounds like an explanation for this "effect".
>
> I have one more question concerning size. The following table
> create table test(val0 unsigned smallint,
> ... ... ... .. ..., val7 unsigned smallint,
> bulk bytea);
>
> needs app. 100 Bytes per row even though the user data only uses
> 8*2Bytes (vali) + 40Bytes (the sizye of my bytea). Does that mean, that
> the system needs 46 bytes for internal represantationsor is the
> alignment factor a again a source of this "blowup"?

Well, it's in the FAQ under "How much database disk space is required
to store data from a typical text file?" but the per tuple overhead is
between 36 and 44 bytes. depends a bit on the version.

Have a nice day,
--
Martijn van Oosterhout <kleptog(at)svana(dot)org> http://svana.org/kleptog/
> Patent. n. Genius is 5% inspiration and 95% perspiration. A patent is a
> tool for doing 5% of the work and then sitting around waiting for someone
> else to do the other 95% so you can sue them.