Wireless Beowulf?

Mark Hahn hahn at physics.mcmaster.ca
Mon Nov 25 12:56:50 PST 2002


> wireless is a shared media so scaleability is poor...

yes, but it would be so cool!  especially if each node was 
solar powered, and thus entirely wireless.  yeah, then we could
shrink them to a single, flyspec-sized chip, and also use much 
higher frequencies (light).  oops, sorry, that was Vinge's 
"a deepness in the sky" ;)

seriously, don't these 802.11* protocols generally have a fairly
low limit to the number of channels used in one proximity?
I seem to recall 3 for 802.11a, which would lead to just a handful
of MB/s aggregate bandwidth, no?  and I imagine there could be some
real propogation issues if you had multiple rows of racks ;)

> > Has anyone researched using the wireless 11a (or soon to be released 11g)
> > protocol as an idea for simplistic beowulf design? I know it won't be as
> > fast as fiber, but rather than having 5,000+ wires hanging everywhere,
> > perhaps it could be a viable idea in a closed-room location?




More information about the Beowulf mailing list