On Mon, 23 Feb 2004 17:40:24 -0500
Patrick Finnegan <pat_at_computer-refuge.org> wrote:
> Likewise, we've seen Supercomputing go from very centralized machines
> - vector processors like the Cray-1, Astronautics ZS-1, CDC Cyber 205,
> etc. - to clusters of machines such as the ever-present (typically
> Linux-based) clusters of machines that are fairly symmetric and don't
> have a whole lot of glue, other than shared filesystems and network
> switches, between them. This stage is like the outflux of PCs in the
> 80s to early 90s. This also includes other things like IBM's SP-type
> systems, Sequent (now IBM) NUMA-Q stuff, the Sun F6800's we've now got
> at work with their "Fire Link" intereconnect, etc.
I have to disagree here. There is an important difference between a
"Linux cluster supercomuter" and a NUMA machine like e.g. the SGI Origin
/ Onyx: A Linux cluster consists of independent, loosely coupled
machines with independent main memory, each runing it own operating
system instance. A NUMA machine consists of tight coupled compute / IO
units, runing a single operating system instance on all CPUs and a
global shared memory. Big NUMA machines are IMHO supercomuters, where
Linux PeeCee clusters are not.
--
tsch??,
Jochen
Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/
Received on Tue Feb 24 2004 - 03:00:48 GMT