Subject: Re: Installation on a VAXStation 4000/60
To: None <john@sixgirls.org>
From: Carl Lowenstein <cdl@mpl.ucsd.edu>
List: port-vax
Date: 05/13/2001 12:45:17
> From port-vax-owner-clowenst=ucsd.edu@netbsd.org Sat May 12 08:32 PDT 2001
> Delivered-To: port-vax@netbsd.org
> Date: Sat, 12 May 2001 10:32:02 -0500
> From: Brian Chase <bdc@world.std.com>
> To: John <john@sixgirls.org>
> cc: NetBSD/vax Mailing List <port-vax@netbsd.org>
> Subject: Re: Installation on a VAXStation 4000/60
> MIME-Version: 1.0
> 
> On Sat, 12 May 2001, John wrote:
> 
> > I just bought my first VAX (VAXStation 4000/60), and I'm very excited
> > about setting it up.
> >
> > Finally, regarding performance: I am currently running NetBSD on a 60 MHz
> > 68060 with about 100 users (about 5-20 are logged in at any one time). ssh
> > is fast enough to not give any noticeable lag on this system, but on a
> > 68040 that I use for testing, I do notice that with three or more people
> > logged in at a time, things sometimes get pretty slow.
> >
> > Could anyone tell me if this VAX is going to be fast enough to handle a
> > fair number of users (perhaps up to 10) connecting via ssh? It'll probably
> > only be running email, but I'd like to know if my users are going to be
> > happy with the machine the way it is, or if I need to have them ssh to
> > another machine and connect via the local (secure) segment to the VAX...
> 
> With just ten users connected simultaneously?  By default the GENERIC
> kernel for VAX is setup to allow 8 users, so you'll want to bump up that
> number and recompile it.  The VS4000/60 is fairly zippy in VAX terms at 12
> VUPs, so I think you'll be okay as long as you have enough RAM.  I'd think
> that you'd want 32MB as a bare minimum.  Given the computational overhead
> of the SSH connections you may end up being getting a little compute
> bound.

Comment on CPU speed.  In the old SPECmark, which is approximately the
same as VUP, a 33MHz NeXTstation (68040) scores about 25.  So you may
well find that your VS4000/60 is a bit on the slow side compared to the
68060.

    carl

        carl lowenstein         marine physical lab     u.c. san diego
        {decvax|ucbvax} !ucsd!mpl!cdl                 cdl@mpl.ucsd.edu
                                                  clowenstein@ucsd.edu