[comp.sys.sequent] Sequent S81 running Dynix

caanders@sactoh0.UUCP (Chris A. Anderson) (03/07/89)

Hello,

My company is considering the purchase of a fairly large (to me,
anyway) computer in the near future.  One of the systems under 
consideration is a Sequent S81 running Dynix.  The model that
they've quoted has about 8 Gb of disk storage and two 386 processors.
They're claiming 8 mip performance.

Could anyone with any experience with Sequents in general and the 
S81 in particular give me some information about the company, the 
OS, and the hardware?

What I'm looking for is System V compatibilty, reliabilty, support(!),
general views on the company, etc.  Anything you can give me would
be nice. We're primarily a System V shop, with little or no BSD
experience, so portability is an issue as well.  We make use of large
data base files ( >250 Mb ), otherwise our applications are pretty 
straight forward.  We do plan to get into image processing later on,
so compatibility with the rest of the world is an issue as well.. 

Thanks in Advance!

Chris Anderson                    email: ...!pacbell!sactoh0!caanders
QMA, Inc.                         uucp:  ...!csusac!fenris 

david@horizon.KAI.COM (03/10/89)

Well, we don't have any S81's, but we do have an S27 with six 386's,
16M memory; as well as a B8 with six ns32k processors 12M memory.  Our
usual load on the S27 is a buch of editors, a few compiles (~4M
virtual memory), possibly a link or two (of a 12M program) and one or
two large (~12M) executables running. Response time is usually very
good. Even when we do a recompile (5 compiles at a time), response
time is still good.

Our system manager had an interesting answer to a survey he filled out
concerning running diagnostics on our Sequent machines. 

How often do you run diagnostics:
	a:  every week
	b:  every month
	c:  when there are problems
	d:  never
He answered c and d.

As far as I can remember, we've only had one hardware problem in the
three+ years we've had the Balance, and no problems yet with the
Symmetry. From what I've read in this news group, there are many
others with similar stories. After reading the story in the January
1989 Electronics about Sequent, it looks like they will be around for
a while.

Disclamer: I wish I worked for Sequent. 8^)

David Nelson               UUCP:      ...!uunet!uiucuxc!kailand!dnelson
Kuck and Associates, Inc.  Internet:  dnelson@kai.com 
1906 Fox Drive             Arpanet:   dnelson%kai.com@uxc.cso.uiuc.edu
Champaign, Ill. 61820      Optimizing software for supercomputers

rdp@pbseps.UUCP (Richard Perlman) (03/12/89)

We are running a 4 processor Symmetry (small cabinet, is that the
S27 (can recall)?).  All in all it has been quite reliable.  We
started out life as a 10 processor Balance and then upgraded to
Symmetry rev-A, we just got our Rev. B boards.

We also run diagnostics "when something goes wrong", which in our
case as happened with three sub-systems.  We have had a lot of
annoying problems with the Fujitsu 9-track drives, the heads seem to
get dirty quickly.  Pluss we had some problems with terminations
on the tape controller cables.

We also had to replace the SCED (serial console/ethernet
controller) board recently when it started seeing phantom
interupts from the console when booting the system.  I kept
changing the console baud rate.

We are now beginning to see what looks like a disk problem on one
Swallow drive after 3 years of steady use.

Hot-line service is good, used to be GREAT.  Major complaint,
System V seems to be only half-heartedly implemented.  There are
problems with term i/o and some simple things (news) just don't
work at all.



-- 
Richard Perlman * pbseps!rdp@PacBell.COM || {ames,sun,att}!pacbell!pbseps!rdp
180 New Montgomery St. rm 602,  San Francisco, CA  94105  |*| 1(415) 545-0233

rich@sendai.ann-arbor.mi.us (K. Richard Magill) (03/14/89)

In article <368@pbseps.UUCP> rdp@pbseps.PacBell.COM (Richard Perlman) writes:
>System V seems to be only half-heartedly implemented.  There are
>problems with term i/o and some simple things (news) just don't
>work at all.

What news?  Oxtrap has been running news for almost 2 years now.
-- 
rich.

rdp@pbseps.UUCP (Richard Perlman) (03/14/89)

In article <455@sendai.ann-arbor.mi.us> rich@sendai.UUCP (K. Richard Magill) writes:
>In article <368@pbseps.UUCP> rdp@pbseps.PacBell.COM (Richard Perlman) writes:
>>System V seems to be only half-heartedly implemented.  There are
>>problems with term i/o and some simple things (news) just don't
>>work at all.
>
>What news?  Oxtrap has been running news for almost 2 years now.
>-- 
>rich.


News, as in /usr/news (/usr/att/usr/news)?  According to Sequent
AT&Ts news program is "broken."  When I run it nothing happens,
no matter what, ecev with the -a option and lots of articles in
/usr/news.  Am I clear that I'm not talking about rn, readnews or
vnews here?  If it works, then that's great, now how do I get it
to work on our system.

While I'm on problems.. (Not Sequent's, this one) we have ksh on
our Symmetry (3.0.12) and it works fine as an interactive shell,
but as a scripting language it hangs at times for no apparent
reason.  We havn't really looked to hard to fix it, especcialy
since we just got ksh-'88.  Anyone else have this problem?


-- 
Richard Perlman * pbseps!rdp@PacBell.COM || {ames,sun,att}!pacbell!pbseps!rdp
180 New Montgomery St. rm 602,  San Francisco, CA  94105  |*| 1(415) 545-0233

rich@sendai.ann-arbor.mi.us (K. Richard Magill) (03/19/89)

In article <376@pbseps.UUCP> rdp@pbseps.PacBell.COM (Richard Perlman) writes:
>News, as in /usr/news (/usr/att/usr/news)?  According to Sequent
>AT&Ts news program is "broken."

No offense intended but I never have gotten the SysV news to work.
But then, I never tried long because the next minute I replaced it
with Bnews so...

>While I'm on problems.. (Not Sequent's, this one) we have ksh on
>our Symmetry (3.0.12) and it works fine as an interactive shell,
>but as a scripting language it hangs at times for no apparent
>reason.  We havn't really looked to hard to fix it, especcialy
>since we just got ksh-'88.  Anyone else have this problem?

Haven't tried ksh-88 but previously we saw no problems.  We usually
write scripts in sh though...


-- 
rich.