[Info-vax] VSI has a new CEO
Bill Gunshannon
bill.gunshannon at gmail.com
Thu Aug 5 09:43:40 EDT 2021
On 8/5/21 9:29 AM, Arne Vajhøj wrote:
> On 8/5/2021 9:16 AM, Bill Gunshannon wrote:
>> On 8/5/21 8:38 AM, Jan-Erik Söderholm wrote:
>>> "Everyone" has a laptop supporting VirtualBox.
>>>
>>> "VirtualBox runs on Windows, Linux, Macintosh, and Solaris hosts..."
>>
>> Two things:
>>
>> 1. Unless it has changed considerably from the last time I ran it for
>> anything, VirtualBox is unsuited for any kind of production environment.
>> And development of software intended for use in a production environment
>> is also production.
>
> Not really.
>
> Development environments are quite different from production environments.
>
> Windows 7 or 10 vs Windows Server 2016 or 2019 is very clear.
>
> Ubuntu or whatever vs RHEL or CentOS / Rocky Linux is also somewhat
> distinct.
You are looking at it from the standpoint of features I am looking
at from the standpoint reliability and usability. As you might
imagine at this point, I was not impressed with VirtualBox as
compared to things like VMWare and Hyper-V.
>
>> 2, Many people here have expressed their desire to work in a VMS
>> environment. Not Linux, not Mac and certainly not Windows. For
>> one thing, it requires the acquisition of knowledge they may not
>> wish to have. This could be a deal breaker. Management may see
>> it as: "If I have to run Linux in order to run VMS, why am I
>> running VMS?"
>
> If you develop for or study VMS then you certainly need
> VMS.
>
> But most people will have something else than VMS running
> today.
>
> We all know about one exception. But I think that is an exception.
My grandson uses ChromeOS. I really don't expect he knows how to
administer it. Running VirtualBox is not in a production environment
is not just a user task.
>
>>> I do not see why clusters and HBVS would not work in an VirtualBox
>>> environment. Not with physical shared disks of course, but over
>>> the network.
>>>
>>> In the PDF from the last Webinar there is a SHOW CLUSTER output
>>> shown having 10 nodes:
>>> 2 nodes with V9.1 (x86 in a VM)
>>> 2 nodes with 8.4-1H1 (older IA64)
>>> 2 nodes with 8.4-2L2 (should be Alpha, I think)
>>> 4 nodes with 8.4-2L3 (latest IA64)
>>>
>>> This might use some other VM environment than VirtualBox, but from
>>> the VMS point of view that should not matter, as far as I understand.
>>
>> And that is another issue. In order to run VMS in a production
>> environment you will now have to also run something like VMWare
>> in a production environment with all the associated costs of doing
>> it.
>
> Pulling in VMWare or similar just to run VMS would certainly
> have significant cost.
>
> But most companies already have that in place: VMWare
> and/or IaaS cloud. And it will be a cost reduction to move
> VMS into the same environment that everything else is
> running on.
Not necessarily, but you are free to continue to believe that.
In the 25 years I worked at the University I watched us go from
an all bare-metal environment to a hybrid with a number of VM
Hypervisors. I had to learn a lot. But, I'm a geek. I enjoyed
playing with new toys. I have worked in places where making a
programmer learn a new language would result in their leaving
instead. Like it or not, VM's are not the ultimate answer to
everyone's desires. I still prefer real hardware to virtualization
when it comes to the VAX and that isn't just with VMS.
bill
More information about the Info-vax
mailing list