[Info-vax] how many people are running X86 variant of OPenVMS in Production?

[Arne Vajhøj]

On 9/29/2023 1:57 PM, Simon Clubley wrote:
> On 2023-09-29, Arne Vajhøj <arne at vajhoej.dk> wrote:
>> On 9/29/2023 8:38 AM, Simon Clubley wrote:
>>>> On Wednesday, September 27, 2023 at 8:53:13?PM UTC-4, David Turner wrote:
>>>>> How many people are running X86 variant of OpenVMS in Production?
>>>>> That is on the virtual machine....
>>>>> Just want to see when I should retire....
>>>
>>> As of today, that number should be zero as x86-64 VMS is nowhere near
>>> ready for running the mission-critical workloads that VMS is used for.
>>
>> VMS x86-64 so far seems very stable.
>>
>> The vast majority of bug reports relates to the native compilers and
>> they are technically still in field test I believe.
> 
> The job/pension question also kicks in, as in manager to minion:
> 
> "Why should I risk my job/pension replacing our existing systems with
> something that isn't even finished yet ??? What possible reason could
> I have for authorising its purchase today instead of in a couple of
> years when it is finished and it has had a chance to prove itself ?"
> 
> Unless you are in a desperate need for some specific reason, and hence
> are willing to take the risk, I can't see an answer to the above that
> will satisfy a manager which isn't emotionally involved in the VMS world.

Everything staying forever as today is usually not an option. Especially
not for someone risk adverse as hardware and software goes EOL and
becomes risks.

So for the risk adverse then a VMS Alpha or VMS Itanium system
will have to be replaced at some point in time.

The questions are:
* replace with a VMS x86-64 system or with a non-VMS system
* when to replace

Cost wise it will be lower migration cost to stay on VMS and lower
migration cost by changing faster (operational cost for x86-64 will
be lower than for Alpha/Itanium).

Risk minimization means to start the testing as soon as
possible and wait going to production until any additional
test (in house or external) is not likely to reduce risk
further in any significant way.

If the system has been tested 3 or 6 or 9 months (depending on
the complexity of the application) and there are no outstanding
issues, then there is no point in continuing waiting.

Arne