Improve dump and restore time

From: Pascal Cohen <pcohen(at)wimba(dot)com>
To: pgsql-general(at)postgresql(dot)org
Subject: Improve dump and restore time
Date: 2008-10-10 15:17:09
Message-ID: 48EF71F5.5050705@wimba.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Hello,
I am studying how to migrate our Production Database which is running
under PG 8.2 and we would like to move to PG 8.3
I have read that the only safe solution is to perform a dump and restore.
Our DB is around 6GB large.
I wanted to have an expectation of the migration duration and performed
it on a less powerful machine than the one we have in Production.
Unfortunately it took very very long time (around 27 hours!).
Are there best practices to reduce the migration time ?
Some kind of questions I try to solve:
- Is it better to use a compressed dump or not ?
- Should I use a kind of trick like reading dump file from network while
restoring on the machine to reduce concurrent I/O and so on

Any advice is welcome.

Thanks!

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Bill Thoen 2008-10-10 15:26:04 Bug In COPY TO?
Previous Message Grzegorz Jaśkiewicz 2008-10-10 14:26:56 Re: Using subquery or creating temp table