Beowulf in a Box

Kragen kragen@pobox.com
Sun, 27 Sep 1998 20:12:39 -0400


On Sun, 27 Sep 1998, Greg Lindahl wrote:
> > My feeling is that the main reason that previous specialist hardware system
> > never really took off is that the programming environment was typically too
> > complex and idiosyncratic, and keeping up with microprocessor development
> > was too difficult.
> 
> The box you describe also will also have a difficult programming
> environment -- it's a "clump", a cluster of multiprocessors, which can
> be much more challenging than a simple cluster.

Well, it should be very easy to get something to *work* on it -- each
of the CPUs will be running a separate instance of NetBSD, and will be
addressable via TCP/IP.  Code that runs over Ethernet TCP/IP ought to
port to it as easily as Informix ported to Linux.

Getting it to work *fast* could conceivably be a more difficult
challenge for particular kinds of code, since all the processors will
have to talk to the outside world through whatever is on a single
host's PCI bus, and routed through that host's routing software.

Is that what you meant?

My feeling is that, since there are already numerous people who are
benefitting from Beowulfs with no more than 40 CPUs, and no better
connectivity than dual Fast Ethernet, this will be enough to satisfy a
lot of needs.  I suspect the lack of floating point will be a bigger
disadvantage.

> Hopefully the vendor
> will be able to continue to keep up with microprocessor development,
> but other boards were based on similar processors which suffered from
> technology failures, such as the transputer.

The vendor is Intel; it's more a matter of whether they want to keep up
with microprocessor development.  They're making noises like they plan
to.

One previous board along these lines used Intel i860s, though, and
didn't take over the world.

Kragen

-- 
<kragen@pobox.com>       Kragen Sitaker     <http://www.pobox.com/~kragen/>
The sages do not believe that making no mistakes is a blessing. They believe, 
rather, that the great virtue of man lies in his ability to correct his 
mistakes and continually make a new man of himself.  -- Wang Yang-Ming