Re: Clustering, mirroriing, or replication?

At each store, there are a number of POS registers that are networked.  If
they enter a ticket at one register, they should be able to go to any other
register to tender the transaction, etc...

We already have mechanisms in place to remotely download and install
software on existing PCs, so no additional outlay of cash would be required
to set up these existing machines with more software.  To install new PCs in
800 stores at the same time would be a nightmare of logistics that we can't
currently contemplate.  We also can't contemplate what nasty things will be
done to us if we go to the executives and say we're going to fix their
problems by forcing the stores to rely on only one machine.  The restaurant
business is not kind to PC hardware.

One option is to have the POS software always write to a local database,
with that change getting replicated to all other active nodes.  Another
option is to have a "master" database, with a failover as backup.  Downside
here is that if both machines fail, then they are dead in the water.  We
can't make money if the registers aren't operational.  However, this is
similar to what happens now (which is why we want to change it.)

Having mucked with postgres replication a little in the last couple of days,
I'm starting to wonder just how long it will take us to develop a good
replication mechanism.  If one is already in place, I'd just like to use
that.  Or at least learn from someone else trying to set up something
similar.


Thanks!

Jon Brisbin
=======================================
Development Programmer
NPC International
(620) 235-7726


-----Original Message-----
From: pgsql-general-owner@postgresql.org
[mailto:pgsql-general-owner@postgresql.org]On Behalf Of Richard Huxton
Sent: Thursday, August 12, 2004 3:32 AM
To: Jon Brisbin
Cc: pgSQL General
Subject: Re: [GENERAL] Clustering, mirroriing, or replication?


Jon Brisbin wrote:
> We're trying to figure out how we can implement a reasonably simple
cluster
> of postgres servers on a private network at our store locations.  The idea
> is to have a group of 2-6 PCs each be able to share data in a replicated
> manner.  Having a separate database server is not an option due to the
> immense cost and logistics (we have over 800 stores), so we need to find a
> way to mirror or replicate data between all the nodes currently up and on
> the network.  Some nodes might be not always be up, so we need a failover
> system.  I'm testing to see if the dbmirror that comes with Postgres will
> work, but I haven't got it completely figured out yet.

I'm talking well out of my experience here, having never tried to set up
800 sets of desktop PCs as replicated shared DB servers, but I can't
believe it's more expensive to put servers in. In my experience, even a
cheap white-box dedicated server is more reliable than a PC being used
for desktop apps.

Assume each user kicks the power cable out of their PC once a year, that
means you'll be dealing with several failovers a day on average.

Assuming you can't get a server for each store (that'd be a minimum of
£400 * 800 = £320,000 in the UK), is there any networking between them?

--
   Richard Huxton
   Archonet Ltd

---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?

               http://archives.postgresql.org