<div>This led me to SUNMOS (OS for parallel processing, Sandia's alternative for the aforementioned intel PSC descendants) <a href="http://en.wikipedia.org/wiki/SUNMOS">http://en.wikipedia.org/wiki/SUNMOS</a>. Sandia's page looks like real progress for inertial containment sustainable thermonuclear fusion, which really wouuld be super duper cool, fuel your zeppelin with seawater, but I can't find latter-day references to SUNMOS. Anybody know what became of it?
</div>
<div>Thanks,</div>
<div>Peter<br><br> </div>
<div><span class="gmail_quote">On 5/23/07, <b class="gmail_sendername">Jim Lux</b> <<a href="mailto:James.P.Lux@jpl.nasa.gov">James.P.Lux@jpl.nasa.gov</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">At 02:20 PM 5/23/2007, Larry Stewart wrote:<br>>Peter St. John wrote:<br>>>Mostly I was thinking of TMC (famous for the animation in Jurassic
<br>>>Park), 1982-1994, from MIT, mostly acquired by Sun; so something<br>>>from CalTech maybe predating that, or competing with it, would be<br>>>very interesting. I'll look, thanks.<br>>>My only artifacts are DOS
3.2 and SVr4 manuals :-)<br>>>Peter<br>>><br>>><br>>The problem with hypercubes is that the number of NICs per node<br>>grows with the machine size. Want to double your machine size?<br>>Then add a NIC to every node you already have. The number of nics,
<br>>or links, or cables, grows O(NlogN) for N nodes. You get a machine<br>>diameter which is log N, which is nice, but there are ways to do<br>>that with a fixed number of NICs per node such as fat trees or the
<br>>Kautz/deBruijn family.<br><br><br>Which is why Intel went to the "8 port switch in each node" sort of<br>architecture with the iPSC/2. The iPSC/1 got icky with having to<br>keep adding ethernet cards into each node. The iPSC/2 went to a
<br>packaging with each node on a single card, with two flavors of nodes,<br>depending on whether it had a numeric/vector coprocessor or not. You<br>could put 128 nodes in a single rack, as I recall.<br><br>(iPSC => intel Personal Super Computer)
<br><br><br>>TMC had, among other luminaries, Richard Feynman to help work out<br>>the routing software.<br>><br>>-Larry<br><br>James Lux, P.E.<br>Spacecraft Radio Frequency Subsystems Group<br>Flight Communications Systems Section
<br>Jet Propulsion Laboratory, Mail Stop 161-213<br>4800 Oak Grove Drive<br>Pasadena CA 91109<br>tel: (818)354-2075<br>fax: (818)393-6875<br><br><br></blockquote></div><br>
!DSPAM:4655ac8954436865219710!