Re: A proposal for a General Clustering Framework

[David Brower <David.Brower@oracle.com>]

Jeff Darcy wrote:
> 
> > I have built message based RPC systems that didn't require replies; they
> > were RPC because they completely conformed to CORBA interface definitions
> > and "oneway" semantics.
> 
> If it's one-way in the sense of not waiting for a reply, it's not RPC
> according to any meaningful definition of the term.  

Please tell the OMG that CORBA is not RPC, http://www.omg.org. 

> > I will note that
> > the kernel components of one commercial cluster system I'm working with
> > use RPC interfaces to do membership elections, so there is proof by
> > existance.
> 
> Not so fast, cowboy.  Is this the one-way RPC-that-is-not-RPC of which you
> spoke?  What product is it?  How many nodes does it support?  When you say
> you "work with it" does that mean you're intimately involved with the
> protocol it *really* uses, or do you work with it in some more distant way?

The wording means "if I felt I could say, I would." 

> It's not that I distrust you personally, despite your manner. 

ditto, mr. "work I did that IBM claimed credit for".

> Nobody's saying you should "hand-roll" messages.  Well, OK, I'm not.  ;-)
> While we're quoting earlier messages in the thread, you might notice that my
> very first response said that XML was a great idea *as a representation*.
> What I don't think works so well is tying implementors' hands with a "send a
> request to X, wait for a response from X" communications paradigm that's an
> ill fit for N-way coordination.

Good, we agree on something.

> > If I were -really- religious, I'd say to take one of the free ORBs
> > and use that with a new transport underneath.
> 
> Are any of those free ORBs HA-capable, capable at least of failover if not
> of outright distributed operation?  If not, they're simply not usable for
> many people.

Do you -need- HA on the point-to-point messaging internal to the
infrastructure?   If it doesn't work, you may have a partition and a need
to reconfigure.   Or maybe you can explicitly try another path (object ref
in corba terms). It's hard to  have HA messaging if you don't get a response 
back, so no "oneway" datagram messaging scheme is HA.  It -is- a perfectly 
good debate topic whether the support of multiple paths between nodes should
be exposed or hidden from the layer doing the service message exchange. 
I think sometimes you'll want to take any path, and sometimes you want 
to require -this- path.

> Yes, I know there are people out there interested in non-HA clustering.
> With all due respect, I consider that a pretty different problem.  The
> world's very different when failure is the "normal" case, worth optimizing
> to minimize downtime, rather than a rare case in the shadow of optimizing
> for throughput.

Right.  "Only the failure cases are interesting."  

I like splitting hairs at least as much as the next pedant, but do
we really need a holy war about the definition of "RPC"?  "RPC" is
like "Cluster" -- it means what the speaker wants it to mean, and
no one can honestly say he was wrong.

Hell, I choke thinking of XML-RPC or SOAP as RPC.  Where's the IDL?
Where's the language binding specifications?  They're just message 
formatting standards, and the "O" in SOAP is a lie.

-dB

Linux-cluster: generic cluster infrastructure for Linux
Archive:       http://mail.nl.linux.org/linux-cluster/