[Info-vax] Viability of VMS in the embedded world ?, was: Re: OpenVMS x64 Atom project

[Dave Froble]

On 6/2/2021 2:05 PM, Simon Clubley wrote:
> On 2021-06-01, John Dallman <jgd at cix.co.uk> wrote:
>> https://www.youtube.com/watch?v=3H6AJigJnNs
>>
>> Running VMS on small systems for "edge computing," which is given the
>> clearest explanation I've encountered for that term.
>>
>> John
>
> Does anyone really think that this is a viable market for VMS ?

Apparently someone thinks it might be.

> How many times more expensive than the US$300 board will VMS be
> to run on that board in a commercial environment ?

Since pricing for such a use doesn't yet exist, how can anyone answer 
that question?

> Why would someone learn VMS for this instead of just using an
> established bare metal or RTOS alternative ?

Someone who knows VMS and doesn't know your RTOS ...

Someone who wants to use Basic ...

> Will you need to renew the VMS licence every year to keep your
> embedded application running ?

See above concerning cost ...

> This is exactly the kind of application use that VAXELN was designed
> for a generation ago (ie: VMS didn't cut it even back then) and for
> which multiple well-established dedicated RTOS options exist these days
> if you need something larger than a bare metal program.

I could imagine a situation where what VMS can do makes a particular 
task much easier.  It really would depend on several things, including 
the task requirements, what VSI has to do the job, and such.

> This is an example of the kind of thing that's available for free these days:
>
> https://www.rtems.org/
>
> An example of a technical issue:
>
> If you are doing GPIO operations then you need a dedicated RTOS or
> bare metal program which then talks to a larger system unless you
> are doing something with very low bandwidth that can tolerate a high
> degree of jitter.
>
> What kind of jitter are you going to see if you try driving the GPIO
> pins directly from VMS ?
>
> Even ignoring jitter, then at what kind of rate will you be able
> to drive the GPIO pins directly from VMS ?
>
> You will not be doing hard realtime with this setup.

How often is hard real time actually needed?

How often is a more robust environment needed?


-- 
David Froble                       Tel: 724-529-0450
Dave Froble Enterprises, Inc.      E-Mail: davef at tsoft-inc.com
DFE Ultralights, Inc.
170 Grimplin Road
Vanderbilt, PA  15486