[Info-vax] VMware

[Simon Clubley]

On 2019-12-10, Grant Taylor <gtaylor at tnetconsulting.net> wrote:
>
> When you start talking about failures that can take out multiple 
> buildings in close proximity to each other, you REALLY need an EXTREMELY 
> robust solution.
>

That's what you get with VMS clusters (if you are prepared to pay the
licencing costs.)

Updates across cluster nodes 10s of kilometres apart happen in real time
(not in _near_ real time, but actual real time) and with every VMS
cluster node being in an active configuration, not in some active/passive
configuration.

If something takes out one site without warning then your VMS cluster
works out by itself that the site has gone, automatically drops the
now missing cluster nodes and reconfigures itself to carry on normal
operations without any completed I/Os being lost, even if they were
completed the immediate instant before disaster struck.

It doesn't matter if it's a database I/O or some text file that was
changed with an editor. If you are working with VMS cluster storage,
it's safe even if you were working on the node that just got destroyed.

Unfortunately, the last time I saw prices for this, it cost stupid
money for the clustering licences.

Simon.

-- 
Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.