[Simh] Deathrow OpenVMS SIMH Node..

[Stephen Hoffman]

Triple-architecture clusters of VAX, Alpha and Integrity Itanium  
systems have been tried and found to work at various sites.

In my admittedly somewhat spotty recollection, the OpenVMS Cluster  
V8.3-1H1 Software Product Description (SPD; <http://docs.hp.com/>) has  
softened the wording around the lack of support in this area, too.     
While not supported as far as warranty support or migration support  
and explicitly not supported for production use, apparently HP is  
building some confidence in the operations of these configurations.   
This based on the wording in the current SPD.

Would this be supported by HP?  No.  Not beyond the wording present in  
the OpenVMS Cluster SPD; not without some sort of specific support  
statement from HP.

But will it work well enough for an emulated VAX to join into a  
hobbyist cluster such as Deathrow?  Probably yes.   I see no reason  
not to try it here, and (on no evidence) would tend to expect issues  
arising with the emulation and less so with the clustering.   (There's  
been an oddity or two here.)  Would I run this triple-architecture  
configuration in production?  I'd certainly think about it, if I  
really needed a VAX and had no way to VEST/AEST/translate or to port  
the application code involved.   The obvious acceptance testing - in  
the absence of site-local and application tests or of the readily- 
available collection of excessively creative users commonly found on  
the Deathrow cluster - would be to crank up multiple parallel passes  
of UETP.

But I'd not expect HP to support any problems that might be found.

Stephen Hoffman
HoffmanLabs LLC

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20090310/7709194d/attachment-0002.html>