Обсуждение: dbsize & pg_dump

Good afternoon,

Probably an easy question but why are the file sizes differ so much between these two tools?

For example: 

A backup using pg_dump of our largest DB creates a file 384MB in size

Using the following SQL code utilizing dbsize I get the following:

FIX1=# SELECT D1.pg_size_pretty AS "FIX1_DB_SIZE"

FIX1-# FROM (SELECT pg_size_pretty(database_size('FIX1'))) D1;

 FIX1_DB_SIZE

--------------

 3832 MB

(1 row)

Is it safe to assume that pg_dump does a compression of the data?

Thanks

Tim

Usually a dump is significantly smaller than a live database due to
space taken up by indexes and discarded tuples from MVCC. If it's
significantly smaller you may also want to take a look at your vacuuming
procedure.

But I'm not sure database_size() is.

Jason Minion
jason.minion@sigler.com


________________________________

From: pgsql-admin-owner@postgresql.org
[mailto:pgsql-admin-owner@postgresql.org] On Behalf Of mcelroy, tim
Sent: Tuesday, April 25, 2006 4:06 PM
To: 'pgsql-admin@postgresql.org'
Subject: Re: [ADMIN] dbsize & pg_dump



Please disregard this question.  I'm using pg_dump -F c which compresses
the data a it backs it up.



Tim





-----Original Message-----
From: pgsql-admin-owner@postgresql.org
[mailto:pgsql-admin-owner@postgresql.org]On Behalf Of mcelroy, tim
Sent: Tuesday, April 25, 2006 4:40 PM
To: 'pgsql-admin@postgresql.org'
Subject: [ADMIN] dbsize & pg_dump



Good afternoon,

Probably an easy question but why are the file sizes differ so much
between these two tools?

For example:

A backup using pg_dump of our largest DB creates a file 384MB in size

Using the following SQL code utilizing dbsize I get the following:

FIX1=# SELECT D1.pg_size_pretty AS "FIX1_DB_SIZE"

FIX1-# FROM (SELECT pg_size_pretty(database_size('FIX1'))) D1;

 FIX1_DB_SIZE

--------------

 3832 MB

(1 row)

Is it safe to assume that pg_dump does a compression of the data?

Thanks

Tim

On Tue, Apr 25, 2006 at 04:47:53PM -0500, Jason Minion wrote:
> Usually a dump is significantly smaller than a live database due to
> space taken up by indexes and discarded tuples from MVCC. If it's
> significantly smaller you may also want to take a look at your vacuuming
> procedure.

Between excluding the database overhead (mostly tuple headers),
excluding indexes, and compression, getting a 10x reduction in database
size isn't unexpected. Using pg_dumpall and bzip2 the databases on
http://stats.distributed.net go from 41G down to 2G.
--
Jim C. Nasby, Sr. Engineering Consultant      jnasby@pervasive.com
Pervasive Software      http://pervasive.com    work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf       cell: 512-569-9461