On Sun, Nov 13, 2011 at 10:45 PM, Andy Colson <andy@squeakycode.net> wrote:
> On 11/13/2011 07:51 AM, Gregg Jaskiewicz wrote:
>>
>> pg_dump -Fc already compresses, no need to pipe through gzip
>>
>
> I dont think that'll use two core's if you have 'em. The pipe method will
> use two cores, so it should be faster. (assuming you are not IO bound).
I am likely IO bound. Anyway, what's the right code for the pipe
method? I think the earlier recommendation had a problem as "-Fc"
already does compression.
Is this the right code for the FASTEST possible backup if I don't care
about the size of the dump, all I want is that it's not CPU-intensive
(with the tables I wish excluded) --
BKPFILE=/backup/pg/dbback-${DATA}.sql
pg_dump MYDB -T excludetable1 -T excludetable2 -U MYDB_MYDB | gzip
--fast > ${BKPFILE}
Thanks!