100 mips isn't enough to run VM?  My first systems programmer job in 1984 we
had a 4331 running VM/SP 2 with VSE as a guest under it (which was where we
moved all the applications from our old Univac machine.)  It only had, I
think, 0.4 mips.  That's 400Kips, folks, and 4M of real memory.  Worked fine
until we also got 20 concurrent engineering types on 3178's doing
interactive work, too.  Then we had to upgrade to a 4361 with 8M of real
storage.

"You don't need a parachute to skydive.  You need a parachute to skydive
twice." - motto of the Darwin Society.
Gordon W.Wolfe, Ph.D. The Boeing Company (425)865-5940


-----Original Message-----
From: Dave Myers [mailto:[EMAIL PROTECTED]]
Sent: Friday, August 30, 2002 7:18 AM
To: [EMAIL PROTECTED]
Subject: Re: H50 sandbox


In a message dated 8/29/2002 10:36:43 PM Mountain Daylight Time,
[EMAIL PROTECTED] writes:



Mark, David,

Thanks for the input.
I'm surprised that 100+mips isn't enough to run what I've laid out,
given dynamically weighting for CP and dynamic storage reconfiguration.

I agree that the flexibility of reconfiguration that VM gives you is a good
reason to
configure everything under VM.  And...the learning experience of using guest
lans
and vctcs etc. will be valuable.

I'm not up to speed on VM yet, so three followup questions:
----------------------------------------------------------------------------
--

-------
>I would try not to have all the guests running simultaneously if at all
possible.

1. Will VM give me more flexibility when I carve up the real memory
     to the individual guests  (vs multiple lpars using dynamic storage
reconfiguration).
     Can I add/delete real memory to guests on the fly?

2. Same question for CP ?


>Dedicate two of the adapters to Linux guests and use guest LANs to connect
>VM TCP and the z/OS systems

3.  Are these two Linux guests suggested for the purpose of routing(to the
other guests)
     DMZ, firewall, etc.  as you have mentioned in the past?

Tia
Dave

Reply via email to