SELECT pg_size_pretty(pg_database_size(current_database()))
-> 1198MB

> Making Slony scale like this is a non-starter, I would think.

Can it cascade?:

Master
    |-Regional Master 1
    |    |-Slave1.1
    |    |-Slave1.2
    |    -Slave 1.3
    |-Regional Master 2
    |    |-Slave 2.1
    |    |-Slave 2.2
...

 Dave Stevenson
dave.stevenson at pacbell.net 




________________________________
From: Peter Geoghegan <peter.geoghegan86 at gmail.com>
To: Dave Stevenson <dave.stevenson at pacbell.net>
Cc: Michael Squires <msquires at whitepages.com>; slony1-general at lists.slony.info
Sent: Thu, February 25, 2010 8:44:29 AM
Subject: Re: [Slony1-general] Is Slony for me?

>     Not sure exactly how to measure this. After installing PostgreSQL to a
> new machine, the 'data' directory is 32MB. I then restore my db from backup,
> the data directory is 1.31GB. However, the .backup file from which the db
> was restored is only 91MB.

To determine the size of your database, do this while connected to it:
SELECT pg_size_pretty(pg_database_size(current_database()))

>     Maybe up to a few dozen or so eventually.

What you must understand is that slony's communication overhead with
respect to the number of nodes increases quadratically; Every time you
double the number of nodes, you quadruple the communication overhead.
According to the docs, "Up to a half dozen nodes seems pretty
reasonable". I'm not sure what the record is for number of slony nodes
in a single cluster, but I suspect it is more than 6 but quite a bit
less than several dozen. Making Slony scale like this is a
non-starter, I would think.

Regards,
Peter Geoghegan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20100225/08d5a1f5/attachment.htm