From: arnold@skeeve.com
To: clemc@ccc.com, arnold@skeeve.com
Cc: tuhs@tuhs.org
Subject: Re: [TUHS] Unix, IBM, 370
Date: Sun, 03 Nov 2019 01:05:55 -0600 [thread overview]
Message-ID: <201911030705.xA375twJ013107@freefriends.org> (raw)
In-Reply-To: <CAC20D2Pz5d4n3ZDhP85d0dX97jLUi=kDKDKdNMZdw2D32Mk72w@mail.gmail.com>
Thaks Clem.
Clem Cole <clemc@ccc.com> wrote:
> On Thu, Oct 31, 2019 at 7:11 AM <arnold@skeeve.com> wrote:
>
> > Tom,
> >
> > Thanks.
> >
> > AIX/370 existed and I *think* would boot on bare metal instead of running
> > on top of VM. I don't know what, if any, relationship it had to the
> > Locus work. (In the late '80s I worked at a university computing center
> > with VMS, Suns, and IBM gear; so I'm recalling what I heard. I never
> > actually saw AIX/370 running.)
> >
> AIX/370 and AIX/386 were done for IBM under contract by Locus Computing
> Corporation (a.k.a. LCC)
> <https://en.wikipedia.org/wiki/Locus_Computing_Corporation> . And yes,
> most customers that I knew ran it bare metal.
Glad to know that I remembered correctly.
In the early 90s I worked teaching multi-vendor Unix courses. One
frustration was that AIX on the 370 and AIX on the PS/2 were essentially
the same as each other but very different from AIX on the RS/6000
machines. A co-worker and I wrote a short essay about if IBM made
cooking equipment:
The IBM Industrial Furnace and the IBM camping stove
would be almost, but not quite entirely, totally different
from the IBM Home Oven.
Or something like that. I can't find the original.
> Because of TCF (Transparent Computing Facility), PS/2 based PC were
> clustered with the 370s, under a single system image (i.e. up to 32
> processors of any time, looked to the world like a single processor). The
> OS looked at the binary and found a properly provisioned system in the
> cluster to execute it. So you could have require option hardware that only
> one node might have, and the process would be migrated to that node. It
> also meant nodes could and be added and removed dynamically.
Very cool.
> The ideas were recreated as 14 different technologies called Transparent
> Network Computing (TNC) that would end up in the FOSS community and added
> to Linux 2x kernel as: OpenSSI <https://en.wikipedia.org/wiki/OpenSSI>
Am I wrong, or does nobody actually use this today? The opessi.org
home page link from Wikipedia just seems to hang. And the files on the
SourceForge page are 5 years old.
Thanks,
Arnold
next prev parent reply other threads:[~2019-11-03 7:06 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-29 4:04 Warren Toomey
2019-10-29 5:07 ` Warner Losh
2019-10-29 5:19 ` Adam Thornton
2019-10-29 7:14 ` SPC
2019-10-29 15:10 ` Warner Losh
2019-10-29 15:22 ` SPC
2019-11-05 4:12 ` Grant Taylor via TUHS
2019-10-31 3:56 ` Tom Lyon
2019-10-31 4:16 ` Larry McVoy
2019-10-31 7:51 ` arnold
2019-10-31 13:51 ` Tom Lyon
2019-10-31 14:10 ` arnold
2019-10-31 14:22 ` Larry McVoy
2019-10-31 14:24 ` SPC
2019-10-31 15:31 ` Charles H Sauer
2019-11-01 16:52 ` Clem Cole
2019-11-03 7:05 ` arnold [this message]
2019-11-03 21:16 ` Clem Cole
2019-11-04 14:43 ` arnold
2019-11-05 14:15 ` Clem Cole
2019-10-31 15:10 ` Heinz Lycklama
2019-11-01 16:40 ` Clem Cole
2019-10-31 8:09 ` SPC
2019-10-31 15:12 ` Warner Losh
2019-11-03 1:02 ` Kevin Bowling
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201911030705.xA375twJ013107@freefriends.org \
--to=arnold@skeeve.com \
--cc=clemc@ccc.com \
--cc=tuhs@tuhs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).