Re: Runtime dependency from size of a bytea field - Mailing list pgsql-performance

From Sander, Ingo (NSN - DE/Munich)
Subject Re: Runtime dependency from size of a bytea field
Date
Msg-id 9EB22E4572ECF74AAFEAE743C74D26B203A875F9@DEMUEXC005.nsn-intra.net
Whole thread Raw
In response to Re: Runtime dependency from size of a bytea field  (Craig Ringer <craig@postnewspapers.com.au>)
Responses Re: Runtime dependency from size of a bytea field
List pgsql-performance
Changing of the storage method ( alter table bytea_demo Alter part1 Set
storage EXTERNAL)
or the increasing of the BLOCK_SIZE (new compilation of the code with
--with-blocksize=32) change the behaviour.

Ingo Sander


-----Original Message-----
From: ext Craig Ringer [mailto:craig@postnewspapers.com.au]
Sent: Wednesday, October 06, 2010 3:24 AM
To: Merlin Moncure
Cc: Sander, Ingo (NSN - DE/Munich); pgsql-performance@postgresql.org
Subject: Re: [PERFORM] Runtime dependency from size of a bytea field

On 10/06/2010 12:11 AM, Merlin Moncure wrote:

> Probably you are hitting toast threshold and running into compression.
>   compression you can disable, but toast you cannot (short of
> recompiling with higher blocksz).

For the OP's reference:

http://www.postgresql.org/docs/current/static/storage-toast.html
http://www.postgresql.org/docs/current/static/sql-altertable.html

While (I think) PLAIN storage could be used, the inability to span rows
over blocks means you would't get over 8k anyway.

--
Craig Ringer

pgsql-performance by date:

Previous
From: Craig Ringer
Date:
Subject: Re: Runtime dependency from size of a bytea field
Next
From: Merlin Moncure
Date:
Subject: Re: Runtime dependency from size of a bytea field