Scientists Claim They’ve Built a Computer That Never Crashes

[kyrathaba]

http://gizmodo.com/5984493/scientists-claim-theyve-built-a-computer-that-never-crashes

[TaoPhoenix]

You know this is ITCHING for hacking.  

Quote:'"Just don't expect it on your desktop any time soon. "

What T.H. good is that if we can't have it?

[f0dder]

Sounds... convoluted. And one ting is "crashes", another your standard run-of-the-mill logic bugs - nothing can really do anything about those.

The executive overview doesn't really give much info, anyway. But it sounds like something that's going to be hard to make reach the speeds we're currently seeing - lots of duplicated units (that might be idle when not selected by the RNG, and thus wasting expensive silicon?), how is the data transfered back and forth, etc.

Also:

Even when it feels like your computer is running all your software at the same time, it is just pretending to do that, flicking its attention very quickly between each program," Bentley says.

...has Bentley been living under a rock for the last, dunno, 10 years? We've had multicore machines for quite a while. Sure, each core runs one step at a time, but cores do run in parallel :-)

Might be an interesting idea, but even in the newscientist story that gizmodo links to, there's not a lot of information. How is the system different from clock-synchronized failover systems that have also been around for quite a while (albeit in the pretty high-end of computing)?

[Renegade]

My initial thought was, oh! They've re-discovered the abacus~!

It would be nice though... Still, I'm sure that NVIDIA and ATI could manage to crash it. They have years of experience!

[f0dder]

It would be nice though... Still, I'm sure that NVIDIA and ATI could manage to crash it. They have years of experience!

To be fair, I don't recall seeing an nvidia driver *crash* (as in BSOD, not "things went flaky, screen flickered, and things were restored") since Vista, when Microsoft yanked big parts of the graphic driver stack back out from kernelmode to usermode. Used to be that way in the early NT days for über security, but most stuff was moved kernelmode in... NT4?... for performance reasons - CPUs were slow back then. Linux is finally catching up on (the right) split between kernel- and usermode, and OSX still has too much kernel (remember the Chrome-can-crash-OSX debacle last year? That was a mix of small bug in chrome, bigger bug in Intel video drivers, and ugly OSX architecture :-)).

Also, graphics drivers are insane these days, they're mini operating systems. Shader compilers, fragment schedulers and sundry.

[KynloStephen66515]

Give it to my mum, for 2 hours...it will crash...I guarantee it.

[Renegade]

Give it to my mum, for 2 hours...it will crash...I guarantee it.

Crash how? My guess is that she'd just get really frustrated with the thing, and living on the 18th floor of the apartment building, the fall would produce a spectacular crash~!

How close am I on a scale of ground floor to penthouse?

[f0dder]

Give it to my mum, for 2 hours...it will crash...I guarantee it.

In seriousness, this is the kind of thing that would be a big scientific breakthrough - securing against end-users. I am going to be very impressed when that happens.

But... mumblegrumble... unintelligiblemumblings... yes, and the quantum flux proves that our Schrödinger devices purrform as promised, the Heisenbox model is proving it's worth; the relativistic probability models are within safe parameters. We are all going to enjoy it down here.

[Rover]

Somehow if idiots claim to be Scientists, they have credibility.  Calling BS. 

Sorry folks, I'm in a mood.

[kyrathaba]

Still, I'm sure that NVIDIA and ATI could manage to crash it. They have years of experience!

LOL! 

[40hz]

Sounds very much like the approach developed for the Linda "coordination language" developed by David Gerlertner and discussed in his book Mirror Worlds.

Linda-like systems are parallel and distributed programming models that use unstructured collections of tuples as a communication mechanism between different processes.

Sometimes the more a concept appears to be new, the more it's just more of the same.

--------------

It also depends on how you're defining a "crash" or a "computer." And at what level of abstraction or virtualization. Hypervisors routinely stay up even when a hosted VM locks up. So does that also count as a computer that never crashes?