[Info-vax] Kermit on Itanium VMS

[Main, Kerry]

> -----Original Message-----
> From: info-vax-bounces at rbnsn.com [mailto:info-vax-bounces at rbnsn.com] On
> Behalf Of Michael Kraemer
> Sent: January-02-10 5:55 AM
> To: info-vax at rbnsn.com
> Subject: Re: [Info-vax] Kermit on Itanium VMS
> 
> Ken Fairfield schrieb:
> > On Dec 29, 1:05 pm, Michael Kraemer <M.Krae... at gsi.de> wrote:
> >
> >>glen herrmannsfeldt schrieb:
> >>
> >>
> >>>In comp.os.vms Frank da Cruz <f... at panix.com> wrote:
> >>>(snip)
> >>
> >>>>VMS is one of the few surviving OS's that pays serious attention to
> the
> >>>>now deprecated if not forgoteen concepts of stability and backwards
> >>>>compatibility from one release to another, so any VMS / IA64 binary
> that
> >>>>you find in the archive:
> >>
> >>>As I understand it, z/OS will still run OS/360 load modules, such
> >>>as the PL/I compiler written and assembled over 40 years ago.
> >>>I don't believe the VMS currently can run any binaries over 40
> >>>years old.
> >>
> >>I wonder if it can run binaries 20 years young.
> >>Or would a VAX binary of 1989 run on current Itanic hardware,
> >>out-of-the-box?
> >
> >
> > Assuming that's a serious question (is it? really?),
> 
> well, yes, at least partly.
> The owners of VMS could have made the various
> transitions as smooth as Apple, for example.
> Invoking an extra translator is not as "out-of-the-box"
> as the z/OS example, IMHO.
> So VAX compatibility was "broken" after 14 years, in 1992,
> when Alpha appeared and one had to switch if one wanted
> more powerful VMS equipment. Alpha compatibility was
> broken after another 14 years, in 2006 (latest),
> when more powerful VMS equipment had to be Itanic based.
> If you look at the competition, PA-RISC lasted about 20 years,
> Sparc more than 20 years, and Power turns 20 this year
> and may still have left a few more years of life.
> In this sense VMS is one of the least backward compatible
> platforms out there.
> 

You are confusing HW platforms with OS version compatibility.

One expects HW platforms will change over time as that is to be expected.

What most Customers expect is that the transition from the old platform
to the new platform will be as smooth as possible.

To that extent, OpenVMS has succeeded exceptionally well. 

As an example, what other OS allows all of its supported HW platforms to 
actively participate in a clustered configuration e.g. VAX (32bit), Alpha
(64bit RISC) and IA64 (64bit CISC)? Yes, App's that might have platform 
specific functions do need to be re-visited, but RMS is RMS and being able to 
share files in an active-active cluster across 3 different HW platforms is a 
pretty impressive accomplishment.

You mentioned z/OS as an another example of good App backward compatibility, 
(and I agree btw), but it also has its compatibility considerations:
http://en.wikipedia.org/wiki/Z/OS

Extract:
"From its inception z/OS has supported tri-modal addressing (24-bit, 31-bit, 
and 64-bit). Up through Version 1.5, z/OS itself could start in either 31-bit 
ESA/390 or 64-bit z/Architecture mode, so it could function on older hardware, 
albeit without supporting 64-bit applications. (Only newer z/Architecture 
hardware manufactured starting in the year 2000 can run 64-bit code.) IBM 
support for z/OS 1.5 ended on March 31, 2007. Now z/OS is only supported on 
z/Architecture mainframes and only starts in 64-bit mode. Application 
programmers can still use any addressing mode, all applications regardless of 
their addressing mode(s) can coexist without modification, and IBM maintains 
an unwavering commitment to tri-modal backward compatibility. However, 
increasing numbers of middleware products and applications, such as DB2 
Version 8 and above, now exploit 64-bit addressing."


Regards,

Kerry Main
Senior Consultant
HP Services Canada
Voice: 613-797-4937
Fax: 613-591-4477
kerryDOTmainAThpDOTcom
(remove the DOT's and AT)

OpenVMS - the secure, multi-site OS that simply works