[Info-vax] Alpha technology lives on in the Top 500 supercomputers?

[johnwallace4 at yahoo.co.uk]

On Sunday, 5 June 2016 07:36:23 UTC+1, Paul Sture  wrote:
> "An Alpha was back in the Top 500 list, in this millennium?"
> 
> <http://labs.hoffmanlabs.com/node/1932>
> 
> Do read the links there. Here's a taster from "Part I":
> 
> "The problem was that Compaq in year 2001 was still too valuable for the
> acquisition – so let’s reduce the value! Since the chip fab was already
> sold to Intel, how about killing off the Alpha itself..."
> 
> I always maintained that the Alphacide was primarily a business decision.
> 
> -- 
> There are two hard things in computer science, and they are cache invalidation,
> naming, and off-by-one errors.

I'd be inclined to treat the ExtremeTech article as "entertainment 
value only" (e.g. confusion between MIPS and Alpha architectures and
Chinese implementations thereof).

The Novakovic ones sound far more plausible, organisationally and
technically. Samsung were taking Alpha chips and systems quite
seriously at one time (e.g. a few readers here may remember the 
AlphaPowered program and associated publicity). Then Samsung lost
interest, along with most other folk. Who wants to fight against
Intel, when it's clear that Intel's bottomless moneypit is going
to ensure that IA64 becomes the industry standard 64bit architecture,
despite the obvious VLIW technical limitations documented in e.g.
https://courses.cs.washington.edu/courses/cse548/05wi/files/Alpha-IA64-Comparison.pdf

But apparently the Chinese carried on with Alpha? Interesting, in
a way.