[MUD-Dev] Re: TECH: Distributed Muds

[Derek Snider]

According to J C Lawrence...
> Derek Snider <derek at idirect.com> wrote:

>> If you are concerned about wasting CPU power, you might want to
>> consider setting up a server cluster (ie: Linux Beowulf Cluster) so
>> that you can effectively have one huge mega-server that distributes
>> tasks evenly over all CPUs.

>> I'm not certain that this is the best solution for high-speed
>> real-time interactive games, but with the proper configuration,
>> equipment and software, it just might be.
 
> Unfortunately shared world systems tend to have both high
> transactional rates and large shared working sets, which make them
> specifically unsuited to Beowulf setups.  Beowulf makes an excellent
> compute farm for cleanly divisible problems with long run times --
> which is a very different problem space.

That's why I said I wasn't sure that a "Beowulf Cluster" was the best
solution, but that maybe with the proper configuration and software
(where by I meant changing the way the system handles IPC data) it
could be made to work.

You'd probably have to build a system where you take advantage of all
the machines being in very close proximity (ie: on the same rack, or
in the same case) and implement some sort of bus-sharing system rather
than using conventional networking technologies.

What would really be useful would be a modular super-computer design
where you could easily "plug-in" more CPUs, RAM, hard-drives and
network cards.  I know that sounds a little far-fetched, but I seem to
recall hearing bits and pieces about such technologies emerging.

If properly designed and implemented, you could have a system that
could be upgraded constantly without rebooting.  If a CPU fails, it
will be flagged as bad, a little red light will go on, and an operator
can hot-swap it for a new one.  Same goes with hard-drives, memory and
network cards.  If you want to replace an old component with a newer
model, all you do is hot-swap it for a new one.  It might be a good
idea to have it set up so that you press a little button on the
component to let the system know to deactivate it before you remove
it.

I know this is straying off topic of "distributed muds", but if we had
the right hardware, we could achieve a system where we could run
everything off of one huge super-computer, and if it started to get
over-loaded, we could add more cpu/memory/storage/bandwidth to it on
the fly.

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev