Exp/Imp data with blobs

From: Alexander Burbello <burbello(at)yahoo(dot)com(dot)br>
To: pgsql-general(at)postgresql(dot)org
Subject: Exp/Imp data with blobs
Date: 2011-11-10 11:25:42
Message-ID: CAJcRiCXRaEytinG1bu=kykhwK2HtH_Ts3HONtTKDKErrHCPO_w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin pgsql-general

Hi,

In one db that I have, there are a few columns that are blob datatype.
This db has around 200MB of data today and as it a development db yet,
so I am replicating data to another db for testing purposes using
pg_dump and pg_restore.

To export the data it is pretty fast, about 3~4 minutes, that means acceptable.
However, when I import this data to another db (even on the same
machine) it takes around 4 hours to perform the pg_restore. I see
during the process that spend most of the time importing the blob
records.
So I stopped to think if I need to adjust this db with different
parameters, or if this behavior is already expected when working with
blobs.
Does anyone have suggestions how can I tune this process??

Here is the basic info about my env.
Windows 32;
Postgres 9.1;
shared_buffers = 256M
maintenance_work_mem = 32M

Any other question or doubt, please let me know.

Thank you in advance.
Alex

In response to

Browse pgsql-admin by date

  From Date Subject
Next Message Karuna Karpe 2011-11-10 14:01:03 error while loading data into table
Previous Message Thorsten Strufe 2011-11-10 11:01:44 lost tablespace

Browse pgsql-general by date

  From Date Subject
Next Message Jay Levitt 2011-11-10 13:04:53 Learning to rephrase equivalent queries?
Previous Message Thomas Kellerer 2011-11-10 10:45:22 Re: count (DISTINCT field) OVER ()