Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!princeton!allegra!ulysses!bellcore!whuxcc!lcuxlm!akgua!mcnc!ncsu!uvacs!edison!steinmetz!davidsen
From: davidsen@steinmetz.UUCP (Davidsen)
Newsgroups: net.unix,net.unix-wizards
Subject: Re: Unix on IBM Mainframes & Clones
Message-ID: <897@kbsvax.steinmetz.UUCP>
Date: Fri, 15-Aug-86 13:23:21 EDT
Article-I.D.: kbsvax.897
Posted: Fri Aug 15 13:23:21 1986
Date-Received: Sun, 17-Aug-86 09:13:02 EDT
References: <39@alberta.UUCP>
Reply-To: davidsen@kbsvax.UUCP (Davidsen)
Organization: General Electric CRD, Schenectady, NY
Lines: 49
Xref: watmath net.unix:8918 net.unix-wizards:19247

In article <39@alberta.UUCP> bjorn@alberta.UUCP (Bjorn R. Bjornsson) writes:
>I would like to here about peoples experiences with
>Amdahl UTS and/or IX/370 under VM.  Indications of
>implementation quality, efficiency, horror stories,
>price (including maintainance if available), etc.,
>is what I'm looking for.

We have been running IX/370 for some time. The following is my personal
experience with it... the kermel is SysV as you would hope. The utilities
*seem* to be the PC/IX utilities, SysIII with additional non-standard
commands and options hung on.

For example, the 'ls' command has none of the SysV options, but there
is an 'li' command which is, in many ways, better. Just don't try to
run a SysV shell script. The vi editor is missing, but there is the 'e'
editor (INed) which is a *fine* editor, but requires the services of a
guru to run with any terminal other than vt100 or PC running PC/IX. On
the good side, EMACS (UniPress) dropped in and runs perfectly.

Programs larger than 100k or so tend to create lots of page faults,
resulting in poor performance in realtime. The mail system is a cobbled
mess of SysIII and INteractive Systems mail. When first delivered it
required the text editor to read mail! My favority 'mailx' is just not
there.

On the good side: really truely FAST!! The C compiler must work pretty
well, most programs run about 10x a VAX 11/780. This is really nice if
you are working with nroff and a large document. The performance is,
again, about 10x a VAX.

uucp performance has been enlightening. For a while 1200 baud was the
fastest speed in terms of effective thruput. This has been more or less
fixed, but it is still about 2:1 slower than going VAX to VAX.

Overall this is a very fast system, but not particularly compatible
with most existing releases. Since a lot of the system stuff has
pathnames starting with 'uts', I wonder about the source of this
version. Perhaps someone running a real UTS system can tell us. There
is also a SysV from AT&T now, but I don't know any details.
-- 
	-bill davidsen

  ihnp4!seismo!rochester!steinmetz!--\
                                       \
                    unirot ------------->---> crdos1!davidsen
                          chinet ------/
         sixhub ---------------------/        (davidsen@ge-crd.ARPA)

"Stupidity, like virtue, is its own reward"