21 Jan
2014
21 Jan
'14
10:23 a.m.
On Jan 20, 2014, at 4:23 PM, Al Kossow wrote:
Just throwing this out to see what other people
think.
I suspect we're at the tail end of the usage life of devices that don't speak
IP.
I'm mostly thinking about networking devices 80's > 00's
So, what needs to be preserved? How much of this does CHM need to do? Is any other
collecting institution already covering this? How much is within scope?
Al, you are probably very far ahead of me, but I have two suggestions:
1) ?bridge? pieces - the Kindergarten example is probably a Dayna EtherPrint-T or Asante
EN/SC type device that allowed a LocalTalk network to bridge to an Ethertalk network. In
terms of long-term, serious restoration/preservation, having one of those available means
having a device that could be (maybe) deconstructed to give information about both types
of network. It also seems pretty helpful to anyone trying to restore hardware on the
less-used (LocalTalk) side.
2) Interconnects used to parallelize processors in supercomputer families - Infiniband and
the like. This may be a hard goal, but it could be that retiring supercomputer sites may
have the stuff available, and recalling the Cray software discussion going on now, this
might be a good time to ask. And, looking far ahead, some characteristics of these
interconnects will be needed to supply the voracious data demands being put on the
internet, so they may serve as prototypes for data center interconnects in widespread use
years from now.
My two cents, HTH.
- Mark
21 Jan
21 Jan
10:48 a.m.
On Jan 21, 2014, at 10:23 AM, Tapley, Mark wrote:
On Jan 20, 2014, at 4:23 PM, Al Kossow wrote:
Agreed, though I should point out that the EN/SC was actually
a SCSI-to-Ethernet converter for SCSI Macs that lacked expansion
slots (e.g. Plus, Classic, not the SE, though I used an EN/SC on
an SE when SE Ethernet cards cost their weight in gold).
LocalTalk itself is pretty well documented, at least. The book
Inside AppleTalk is a pretty great reference for all modes of
AppleTalk, both LocalTalk and otherwise. A lot of the bridges
are pretty dumb, requiring the LocalTalk side and all its
devices to be alive at boot time. I'm of a mind to make my
own that's smarter, but I need to stop pushing more projects on
my stack.
- Dave
Just throwing this out to see what other people
think.
I suspect we're at the tail end of the usage life of devices that don't speak
IP.
I'm mostly thinking about networking devices 80's > 00's
So, what needs to be preserved? How much of this does CHM need to do? Is any other
collecting institution already covering this? How much is within scope?
Al, you are probably very far ahead of me, but I have two suggestions:
1) ?bridge? pieces - the Kindergarten example is probably a Dayna EtherPrint-T or Asante
EN/SC type device that allowed a LocalTalk network to bridge to an Ethertalk network. In
terms of long-term, serious restoration/preservation, having one of those available means
having a device that could be (maybe) deconstructed to give information about both types
of network. It also seems pretty helpful to anyone trying to restore hardware on the
less-used (LocalTalk) side. 
1:54 p.m.
Am 21.01.14 16:23, schrieb Tapley, Mark:
2) Interconnects used to parallelize processors in
supercomputer
families - Infiniband and the like.
I have several boxes of Myrinet stuff
available. Switches, cabels
(copper and optic), PCI cards. The stuff is around 10 to 15 years old,
speed is 1 or 2 GBit/s. (?) I never intended to use that stuff. I just
saved it from the dumpster because of its historical significance. So:
Free to anyone who takes it out of my hands in Kaiserslautern, Germany.
--
tsch??,
Jochen
3840
days inactive
3840
days old
2 comments
3 participants
participants (3)
-
fraveydank＠gmail.com -
jkunz＠unixag-kl.fh-kl.de -
mtapley＠swri.edu