[Info-vax] WHY IS VSI REQUIRING A HYPERVISOR FOR X86 OPENVMS?

[Simon Clubley]

On 2021-01-04, Arne Vajhøj <arne at vajhoej.dk> wrote:
> On 1/4/2021 8:23 AM, Simon Clubley wrote:
>> On 2021-01-01, Jan-Erik Söderholm <jan-erik.soderholm at telia.com> wrote:
>>> With SAN, you can have HBS between datacenters without running VMS-Cluster.
>>> Just shadow multiple disks from different SAN systems at different sites.
>>> Or let the SAN systems hande the mirrowing of your data.
>> 
>> That takes care of the data.
>> 
>> What about the coordination between application and operating system
>> instances at different sites so that another one picks up the load after
>> the original instance fails ?
>> 
>> How is the original instance blocked from resuming operations if it
>> recovers so that you don't have two instances of the application or
>> operating system trying to update the same data at the same time in
>> an uncoordinated way ?
>> 
>> In other words, what is the SAN version of the DLM and the VMS clustering
>> protocols ?
>
> That is not the job of the SAN.
>

That's what I thought but I wasn't 100% sure just in case something had
come along without me hearing about it.

> But application failover is a pretty universal feature also outside
> of VMS, Implementation techniques vary a lot.
>

So it's the same as always in that you also need the additional supporting
infrastructure to provide what you get as standard with VMS clustering,
including being able to detect when a node has only temporarily failed
(in order to prevent everything that comes with that situation) when that
temporarily failed node comes back online.

Simon.

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