Jump to content

680x0 Super Computer?


Recommended Posts

So I saw yesterday cray released a new super computer (I know, I am a sucker for cray. it just fascinates me along with super computers and mainframes in general) but it strikes me with an interesting question.

 

Has anyone built a super computer in the old ages using Motorola 680x0 chips?

I know IBM built a few (I.E. Watson) using PowerPC chips, but has anyone built a super computer/mainframe using the old Motorola 68k chips?

 

 

If anyone has, can I get a reference?

 

 

I think the 68040 was SMP aware, but I could be incorrect about this. I think it would be cool if someone figured out how to do that.

Link to post
Share on other sites

Prior to the 1990s what we'd usually call "supercomputers" (IE, in the "Cray" sense) were usually assembled using parts designed strictly for the job, often fabricated using more exotic fabrication processes than those used in microcomputer-based hardware. And the architectures were usually geared toward achieving one (or a few) very fast execution cores; the mid 80s Cray-2 had "only" four vector cores, but in clock cycle times alone they ran about 20 times faster than the best microprocessors of the era could offer. (And the entire unit together could perform anywhere from a hundred to a thousand times faster than a typical desktop machine on jobs that were well suited to it.) Thus any "super" machine built out of 680x0 CPUs would by definition have to be a massively parallel unit and, well... the slowest/oldest unit in that class using commodity parts I've seen any references to is the Intel iPSC "Hypercube", constructed out of anywhere from 32 to 128 80286+80287 CPU/FPU "nodes". (Many/most MPP machines, like the early Connection Machine CM-1/2s or the competing MasPar units used custom computing elements, not commodity CPUs.) Subsequent iPSC/Paragon machines were built by Intel using 80386s, i860s, and Pentium Pros, but it's debatable if any of the systems built prior to the 1993-era i860 Paragon XP/S really qualified as a Supercomputer.

 

Are we talking about computers which were "super" compared to a typical desktop/minicomputer or genuine "fastest in the world" contenders? Cray's first "commodity CPU supercomputer" (if you include "workstation" CPUs in the commodity definition), the Cray T3D of 1993, was built on the DEC Alpha 21064 which is a ridiculously powerful chip compared to *any* member of the Motorola 680x0 family. (Easily seven to ten times as fast as the 68040 in a typical application.) Later model Connection Machines (and others) used SPARC and of course, SGI had MIPS-based systems that would qualify as "super". ASCI Red, the first supercomputer to crack the one teraflop mark back in 1997, was made out of Pentium II Xeons and advertised by Intel as " the first large scale supercomputer to be built entirely of common commercially available components". It certainly wouldn't surprise me to find out that an experimental 680x0-based massively parallel cluster existed at some point, probably in the form of racks of VMEbus boards filling a room in a college basement, but as a commercial product? Any evidence of that escapes me. If someone has a citation it might be interesting reading.

 

(There are examples of 680x0-based of computers acting as the front end for a cluster composed of other CPUs, such as i860s or Transputers, that are fairly easy to find. But that doesn't count.)

Link to post
Share on other sites

My dad worked with a few Crays when he was in University. They were having problems with it, and he was explaining to me how they debugged it. I'm sure I have it wrong, but the way I understood it is that they first had to drain all the liquid out of the tower into a special container, then directly hooked up some probes to the machine at certain parts. Without the liquid cooling, it could only be run for a fraction of a second or it'd overheat. It had a special mode where it'd step the machine in very short bursts, allowing the guy debugging it to peak at the data. If it wasn't messing up at one part, he take the probes out and stick them somewhere else, then step the machine and analyze.

 

I can't imagine how long it took. I guess they did that song and dance routine until they found the problem, then replaced whatever was damaged.

 

Sure are freaking awesome looking, though. Even til this day, those machines are exactly how I imagined a computer of the future would look. Still do. I wonder if there'd be a market for a miniaturized Cray-1 or Cray-2 case with a modern PC inside? :)

Link to post
Share on other sites
My dad worked with a few Crays when he was in University. They were having problems with it, and he was explaining to me how they debugged it. I'm sure I have it wrong, but the way I understood it is that they first had to drain all the liquid out of the tower into a special container, then directly hooked up some probes to the machine at certain parts. Without the liquid cooling, it could only be run for a fraction of a second or it'd overheat. It had a special mode where it'd step the machine in very short bursts, allowing the guy debugging it to peak at the data. If it wasn't messing up at one part, he take the probes out and stick them somewhere else, then step the machine and analyze.

Fluorinert. It's a non conductive liquid like mineral oil but far more efficient at transferring heat. It's also very expensive and very toxic.

Link to post
Share on other sites
It's been a while, but I was always under the impression it wasn't *that* difficult to set up an environment for cross-compiling under NetBSD? Anything in, I dunno, the mid-tier Pentium III ballpark will still stomp all over your theoretical 8-way 68060.

I know one could cross-compile, I just like working on the native platform. And I like m68k architecturally.

Link to post
Share on other sites

Come to think about it, can anyone actually name a widely commercially-available multiprocessor+shared-memory Motorola 680x0 based machine? There are some mentions on the NetBSD mailing list of an obscure dual-68040 VME CPU board which does have an experimental port but doesn't have functional support for the second CPU. There are some really early/large multiprocessor 80386 based machines like the Sequent Symmetry, but I seriously can't find a Motorola equivalent.

Link to post
Share on other sites
Multi-Processor 68040 NEXT Machine

 

Sometimes people would *call* the NeXT machines "Multi-Processor" because of the standard inclusion of a Motorola 56000 DSP but obviously that doesn't count. (There was also the i860 "NeXTdimension" graphics board option, but likewise, no dice.) Beyond that, well, I certainly haven't seen any credible evidence that a multi-CPU 680x0 NeXT machine ever existed.

 

My brother was most likely wrong.

 

Brothers are always wrong. They're even worse than sisters.

Link to post
Share on other sites

Well, I have no sisters, but my brother is 3.5 Years older than me. He knew enough to immediately buy stock in Apple when Steve Jobs came back. He bought at around $5/share and sold at around $260/share years later. Made a killing, he did. Probably should've held out just a wee bit longer ^_^

 

Either way, it's no big loss on that. My brother was a huge mac head (not sure now, the last mac he bought new was the G4 iMac Lampshade).

 

Now he does Linux/Windows and the G4 is used with software for my twin nieces.

 

 

He still uses his Newton 2100 though!

 

And the iPod he has.

Link to post
Share on other sites

I did the same, except Merrill Lynch screwed me over. Instead of having what would now be over 4000 shares valued at $550+, I have nothing. My brother and I were victims of one of Merrill Lynch's plethora of corrupt traders who charged trade commissions but didn't actually process the trades. So instead of a little over $2 million, I got $80 back from the lawsuit settlement... and nothing else per FTC rules since no shares were actually bought or sold.

 

Hooray for unregulated capitalism. :-/

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...