[Info-vax] Current VMS engineering quality, was: Re: What's VMS up to these

[Johnny Billquist]

On 2012-03-16 22.28, JF Mezei wrote:
> Johnny Billquist wrote:
>
>> is plenty of proof otherwise. What I was saying was that Unix systems
>> with a shared root disk means NFS, and that means everything hangs if
>> the server goes down.
>
>
> Waht about disk arrays ? isn't it possible now to have shared disks
> without NFS because each unix instance thinks the disks are local ?

I've never seen any such thing. How would you expect each Unix instance 
to manage the disk structure with caching and writing the same blocks as 
another instance?

Could you point me to any such product, and maybe I can make a more 
intelligent comment.

The shared disks without a network stack involved is more or less what 
DEC did with CI. Well, the disks are accessible directly by each 
machine, but if you want to access the disks from more than one machine, 
they have to be in a cluster, and they will talk network with each 
other, in addition to talking with the disk, and the network between 
each other is to manage the cluster wide locking and synchronization of 
access.

	Johnny