[Info-vax] VMS on Raspberry Pi 5

[Ahem A Rivet's Shot]

On Thu, 16 Nov 2023 00:47:55 -0500
"56d.1152" <56d.1152 at ztq9.net> wrote:

> On 11/15/23 9:43 AM, Ahem A Rivet's Shot wrote:
> > On Wed, 15 Nov 2023 08:06:00 -0500
> > 	Of course, these environments are built for that purpose and
> > allow the systems developers to assume the Kubernetes infrastructure
> > without having to maintain it. But it is possible to run such workloads
> > in a private data centre - "TrueNAS Scale" is one fairly easy way.
> 
>    Sorry, but McDonalds Corporate, BOA, US Mil, etc ... they
>    are NOT gonna work on Docker or Kubernetes on a bunch of
>    tablets.

	No most corporate use of Kubernetes is on data centres full of
blades, NVMe based SANs, NetApps and Isilon clusters. This is the world
that pays my wages - I know how big it is and who uses it, including
outfits the size you're talking about, many of them are customers of my
current and previous employers.

> IBM didn't pay big for RHEL to run it on laptops,
>    but on its mainframes ......

	Yes they are popular with banks and the like because they *also*
run their old OS-360 stuff without recompiling it, but to anyone who
doesn't need that they are very expensive for little gain. Guess what many
of their customers run in the RHEL environments - yep kubernetes and docker.

>    Decentralized usually = SLOW ... and INSECURE ... despite
>    claims.

	Tell that to Amazon, Google etc. look into the architecture of
Amazon Dynamo and marvel at the way it scales and handles machine failures
and network outages. Mainframes are great up to a point, right up until you
can't get one big enough and then you *need* a scalable solution.
kubernetes is an easy way to get one.

	I was involved in doing it the hard way back in 1990 when the UK
Inland Revenue had a problem their ICL mainframe team declared impossible,
twenty high end 88k boxes and a distributed architecture made it possible.
No mainframe could match the IO bandwidth or CPU power of that solution,
using it effectively took careful design.

	SAAS is huge in the large corporate world, it all runs on
virtual machines and docker containers orchestrated by kubernetes. It's
only insecure if you don't know how to secure it, the most security
sensitive run it all in their own datacentres on hypervisors that they
control. The rest trust the contractual obligations of the companies that
run the data centres (Microsoft, Google and Amazon mostly).

	The big trend in large users these days is micro-services in docker
containers. Instead of an OS image running in the container there is only
the application and support libraries running on bare virtual metal.

-- 
Steve O'Hara-Smith
Odds and Ends at http://www.sohara.org/
Host: Beautiful Theory meet Inconvenient Fact
Obit: Beautiful Theory died today of factual inconsistency