17 Dec
2017
17 Dec
'17
11:53 p.m.
On Sunday, December 17, 2017 at 1:36 PM, Zane Healy wrote:
I'm not sure how AWS keeps track of CPU usage, but a simh VAX instance
running VMS with idling enabled will normally consume very little host system
resources unless it is really doing something. When the network is otherwise
idle, it will wake up briefly once every 10ms to implement a simulated clock
tick and very quickly go back to sleep.
- Mark
On Dec 17,
2017, at 1:25 PM, Mark J. Blair <nf6x at nf6x.net> wrote:
I just had a potentially dangerous thought. I'm already considering learning
about Amazon cloud server stuff and migrating my Wordpress blog from the
web service it presently lives on to my own AWS instance, so I can have more
control over it.
Is anybody doing any DECnet/HECnet related stuff on AWS yet? I wonder if it
might be hard to spin up something like a HECnet portal on an AWS instance
without running up big bills? I'd be willing to burn up to a hundred bucks a
year doing something silly like that, assuming that exposing something like an
OpenVMS 7.3 instance to the public internet isn't a profoundly bad idea.
I think y'all need to talk me away from the edge of the cliff now. I'll leave it
up to you which direction you talk me. :D
That could be interesting, but it could also be very expensive. Isn't some of
the AWS pricing based on how much CPU time you use? The VM I have
running on ESXI is using about 57% of the host CPU. It doesn't matter to me,
but in a situation where you have to pay for the resources you use, I would
think this could be a problem.
BTW, as I recently posted on the SIMH mailing list, I had setup that VM to
throttle, and forgotten about it. When I moved from playing with PDP-11
emulation, to actively running Simh/VAX 24/7, it caused major issues, and a
Raspberry Pi was performing better. Since I changed that, it's my fastest VAX.
:-)