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

[David Froble]

Johnny Billquist wrote:
> On 2012-03-13 15.22, Michael Kraemer wrote:
>> Keith Parris schrieb:
>>
>>> it is the appropriate reaction to a lengthy problem in the network.
>>> VMS is doing exactly the right thing under the circumstances to
>>> protect the data.
>>
>> Crashing an entire workstation cluster due to some network problem
>> can hardly be called "the right thing".
>> At least not of this cluster has to perform serious monitoring
>> tasks and thus has to be "up" all the time.
>> And certainly not when Unix boxen sharing data over the very same
>> network have no such problems.
> 
> True, since in Unix-land, the occasional corruption of data is 
> considered an acceptable tradeoff.
> 
>     Johnny

Joining multiple computers into a cluster is not always a good idea.  Clusters have their 
uses.  But non-cluster also has it's uses.

As for a group of workstations, using one common system disk, and ethernet for their 
cluster interconnect, well, I would call that a "far from robust" configuration.  As 
usual, you're only as strong as your weakest link.

For real-time work, you cannot wait for a cluster to recover from a problem.  Say you're 
capturing real time data.  If you don't trust one system, you'd be better off with two or 
more independent systems, with independent power supplies, and the data feed split to go 
to every system.  Or for process control, and such.

I've never used a cluster.  Never needed the capability a cluster provides.  Very few of 
my customers ever needed a cluster.  VMS is an excellent OS even if you do not use the 
cluster capability.  Great development and operations capabilities.  DLM !!!  Logical 
names !!!