[Info-vax] OT: news from the trenches (re: Solaris)

[Stephen Hoffman]

On 2015-03-12 19:07:31 +0000, <lists at openmailbox.org> said:

>> 
>> IBM is not going to reacquire the fabs, and they're not going to get
>> back into either the x86 or the fab business, and it would not surprise
>> me to see IBM eventually migrate to commodity x86-64 processors even on
>> their mainframes, assuming they don't spin off the z business.
> 
> Unless you're talking about emulation and there are a few choices 
> there, there is no way to run a mainframe on x86 processors nor 
> commodity anything.

IBM has a very long history of emulating previous generations, and they 
acquired a company that is really quite good at some years back; 
Transitive.

> All the software ever written for the IBM mainframe depends on that 
> specific architecture and ISA. From 1964 until today the OS has never 
> been ported to another platform and it won't ever be. To get off the 
> box you have to emulate the hardware.

The applications are translated or emulated, or the applications are 
ported, or it's toast.

> The OS is too big and too complicated and most of all too closely 
> coupled to the hardware (which is why it works so well) to ever be 
> ported.

VAX/VMS won't ever be ported, either.  VMS is far too tied to the VAX, 
after all.


>>> Yeah but you seem to have suggested there's a question whether VMS can  
>>> or should run on cheap hardware given the VMS customer is typically a 
>>> high end customer.
>> 
>> You're suggesting that x86-64 isn't high-end
> 
> That's right!

OK.  Too bad for NSK and for the folks buying Superdome X, then.  Ah, well.

> 
>>> If you think about it I'm not sure porting VMS to Intel is not going to 
>>>  be suicidal. After they see they can run on cheaper hardware, isn't  
>>> cheaper software the next step? And Linux is free and eventually some  
>>> pinhead is not going to approve the budget for that wierd OS nobody  
>>> ever heard of, since Linux is free...
>> 
>> Have you ported OpenVMS applications to Unix?
> 
> No, I have about a week or so experience with OpenVMS.

Might want to try porting some code onto or off of OpenVMS, and gain 
some experience.

> And I'm not talking about porting.

Effectively, you are.  Folks still have to get the business logic and 
the data out of the existing environment, even if this isn't about 
porting the application code.

The folks that are still on OpenVMS have some of the gnaliest problems, 
or they have little interest in porting, or they expect that the port 
will be — as many are — expensive.  Or some combination of those.  In 
many of the environments, hauling the data and the business logic out — 
reformatting and translating and reencoding as necessary — isn't that 
much smaller of a problem than hauling out the code. either.

> I'm talking about customers abandoning/replacing. I think that's going 
> to happen if VMS tries to compete with *NIX and be just
> another Linux/UNIX substitute. I hope they don't try doing that. I 
> would like to see the OS running on top quality (non-Intel) hardware 
> and be a prime product.

So you're talking about porting.  Which you know nothing about.  And 
for all these years where the various boxes have coexisted, and now all 
of a sudden the folks that have VMS applications that they've been 
working with and have probably ported forward once or twice before, and 
that are actually (hypothetically) running on x86 boxes using VSI's 
native V9.0 port, and _now_ they're going to port those applications to 
something else?

Interesting.

This obviously discounting the various folks that are already running 
OpenVMS on x86, using any of the various emulators — they're usually 
doing that because emulation is the least-bad choice available for 
their current predicament.

> 
>> The only bet against Intel and high-volume x86-64 I can see here is
>> with ARM
> 
> Everybody seems to believe Intel is always going to be here. I realize 
> it seems like that today but tomorrow could be different.

True.  But then you're betting on POWER over Intel and AMD and the 
gazillion x86 systems already in use in the installed base, which means 
you're taking a much bigger risk than I'd be comfortable with here, for 
servers and processors higher costs, and for less available hardware 
configurations and suppliers.

Even if Intel and AMD and the entire x86 market implodes, that process 
of implosion will still take longer than that POWER roadmap currently 
lists.

Ah, well.  Time to get back to dealing with all those pesky x86 boxes, 
and to this blasted Alpha and Itanium cluster that's not building the 
code quite right.  Have fun with your bet on POWER, and I hope it works 
on for you.


-- 
Pure Personal Opinion | HoffmanLabs LLC