HP To Launch Linux++ Operating System Next Year

[paupav]

It is obviously Linux with some parts written in the C++ which are added by HP .

[tuxd3v]

... and the MIPS R4000, which IIRC came before Alpha or Itanium.

probably they came first!!

I don't now if the technical approach to design come completely first, but it seems to be one year or something like that first

But in this things intel is a shame!!

They said they own the first 64 bits arch...IA64, wich is developed by HP...And prior to that DEC developed a beautiful RISC processor from alpha arch(which I used in some mainframes running HP-UX, itaniums too), and apparently prior even to that comes MIPS R4000...

More than a decade ago, AMD released the 64bits of the x86 arch...a shame for intel ( because for normal reasons if someone should improve the arch, should be intel...but they are in the "beach under the palm tree's...")l...and AMD called it AMD64...after several names attributes by intel...it ended with the name x86_64..

WTF...the technology is from AMD!!!

So it is better to call it AMD64, to give the credits do the company that developed it...
AMD could do the same to x86 renaming it to AMD32, but AMD have more respect than intel for that!

[bridgman]

Scott Adams always manages to have timely cartoons:

http://dilbert.com/2014-12-18/

[gilboa]

What comes to my mind here, when I see this, is PA-RISC, the Alpha, and the truly innovative HP of the nineties, not the dull, boring, cust-cotting and crappy HP of the noughties. AFAIK HP has a lot of experience developing vaporware for non existant hardware, so, let's see how far this will go...

I still remember the 0.5M$ quad socket Digital Alpha machine I used back in '96. It was an amazing machine (especially compared to the Pentium 60 with a brain-dead FPU).
... But then HP (or actually Compaq) bought Digital and killed it.

- Gilboa

[tuxd3v]

I still remember the 0.5M$ quad socket Digital Alpha machine I used back in '96. It was an amazing machine (especially compared to the Pentium 60 with a brain-dead FPU).
... But then HP (or actually Compaq) bought Digital and killed it.

- Gilboa

yeahh I remember that too...

I administrated one mainframe with it until 2012 ...
The price for parts...is tremendously high :S

but I loved the arch, and I don't understand why HP haven't continued developing it

[liam]

Obviously, I really hope this materializes as neumann/harvard/princeton is ancient, but I don't have much hope. For one thing, I'm not sure memristors are actually physically realizable.
In the meantime, crossbar's rram is finally approaching market. They're well beyond proof of concept and are at the "commercialization phase".
Sure, rram isn't as revolutionary as memristor (limited w cycles), but it provides us with something that is pretty much as fast as ram, and far more durable/scalable than nand/platters.
To take advantage of it the existing os' will need to be rethought a bit. It would completely replace ssds as caching devices, and it's endurance/performance is enough for it to work as something like a hybrid L4/ssd cache. Not really sure that there exists the exact place in the hierarchy for it, yet.

[gilboa]

yeahh I remember that too...

I administrated one mainframe with it until 2012 ...
The price for parts...is tremendously high :S

but I loved the arch, and I don't understand why HP haven't continued developing it

I'd venture an guess that HP knew it didn't have the resources to continue developing the Alpha chip - especially given the huge resources IBM sunk into the Power chips, and Intel sunk into the than-new Xeon P3 CPUs.

- Gilboa

[coder]

Another company who fundamentally "reinvents" something, with the obligatory "quantum leap". It always makes me laugh that they obviously don't understand that a quantum leap is the tiniest leap possible.

A fair point (and not the first time I've heard this particular complaint).

Analogies are rarely perfect. This usage of the term is clearly analogizing a development to the discontinuous nature of a quantum leap - not its magnitude. A better way to describe it would be "revolutionary", as opposed to "evolutionary".

Now, when people talk about a "quantum" of something, they mean the smallest discrete unit of it.

[coder]

I'd venture an guess that HP knew it didn't have the resources to continue developing the Alpha chip - especially given the huge resources IBM sunk into the Power chips, and Intel sunk into the than-new Xeon P3 CPUs.

What HP did was to bet on Intel and partner with them on Itanium. As we've seen, that turned out to be a bad bet, but it explains a few things.

Aside from that, HP already had PA-RISC, which they kept alive for the longest time. So, keeping Alpha [i]and[\i] PA-RISC would've made little sense. Also, I seem to recall that Intel actually acquired the microelectronics unit of DEC. Maybe I'm wrong about that, but if it's true then HP can't really be blamed for Alpha's demise.

Anyway, I think we're beyond one company fundamentally redefining computing. Haven't they learned anything from the rise of ARM, Linux, and industry consortia like JDEC? Even a superior technology won't see widespread adoption, if it's proprietary and surrounded by a wall of patents. This is one of the more convincing reasons I've heard for the failure of IA64, in fact: it was so bogged down with patents that no one could ever build a clone and industry players didn't want to be locked into using chips that only Intel could ever build.

[gilboa]

What HP did was to bet on Intel and partner with them on Itanium. As we've seen, that turned out to be a bad bet, but it explains a few things.

Aside from that, HP already had PA-RISC, which they kept alive for the longest time. So, keeping Alpha [i]and[\i] PA-RISC would've made little sense. Also, I seem to recall that Intel actually acquired the microelectronics unit of DEC. Maybe I'm wrong about that, but if it's true then HP can't really be blamed for Alpha's demise.

Anyway, I think we're beyond one company fundamentally redefining computing. Haven't they learned anything from the rise of ARM, Linux, and industry consortia like JDEC? Even a superior technology won't see widespread adoption, if it's proprietary and surrounded by a wall of patents. This is one of the more convincing reasons I've heard for the failure of IA64, in fact: it was so bogged down with patents that no one could ever build a clone and industry players didn't want to be locked into using chips that only Intel could ever build.

Two comments:
1. PA-RISC was kept on life support before HP killed off Alpha.
Most likely HP was forced to keep it alive long after the release of the Itanium due to contractual / support consideration.
I would imagine that HP chose to co-develop the Itanium as it no longer had the resource to keep developing a CPU core + operating system + application stack alone.

2. The IA-64 didn't fail due to patent issues. It failed due to 3 different reasons:
A. It was late and uber expensive from day one.
B. It had a *very* complex instruction set (VLIW) that left all the heavy lifting to the compilers which in-turn, never got it right. (Intel own ICC was late, and underperforming).
C. Intel planned to kill the x86 line by keeping it in 32bits, forcing the market to far-more-expensive IA64. AMD released the Athlon64, forcing Intel to add 64bit support to the x86_64 line, effectively killing the IA64.

- Gilboa