Hewlett-Packard will take a big step toward shaking up its own troubled business and the entire computing industry next year when it releases an operating system for an exotic new computer.
The company’s research division is working to create a computer HP calls The Machine. It is meant to be the first of a new dynasty of computers that are much more energy-efficient and powerful than current products. HP aims to achieve its goals primarily by using a new kind of computer memory instead of the two types that computers use today. The current approach originated in the 1940s, and the need to shuttle data back and forth between the two types of memory limits performance.
“A model from the beginning of computing has been reflected in everything since, and it is holding us back,” says Kirk Bresniker, chief architect for The Machine. The project is run inside HP Labs and accounts for three-quarters of the 200-person research staff. CEO Meg Whitman has expanded HP’s research spending in support of the project, says Bresniker, though he would not disclose the amount.
Its exciting to see them getting closer to releasing their new storage and networking infrastructure.
>Nothing is going to happen.
I am afraid I cannot share this information with You.
>butthurt from some autistic conspiritard
Try again. The issue is how the memristor is described, not if they actually exist or not. HP has demonstrated working memristors, no one doubts that they actually exist other than utter retards. They're just a type of ReRAM, a very high density and fast one at that.
>comparing completely incomparable things trying to make a point
Its memory. Its faster than everything else out there. They're connecting everything together with fiber optics to eliminate latency. Its compatible with current CPU/GPU architecture.
I'd explain everything on a point by point basis to you, but thats something you do to a 5 year old, and I'm not a kindergarten teacher.
>please explain the entire basis of photonics to me
I don't even know where to begin. I guess its my fault for assuming everyone posting here actually made it out of middle school.
The memristor is great as a successor to flash if they truly have good density and no endurance limit:
but we are already going to photonics for IO. thunderbolt was going to be optical, but they figured out how do squeeze the desired performance out of copper.
Next gen stuff is going to be insane, with new storage tech, optical io, and chip stacking for dram. It will be evolutionary though in that we are still going to be running compiled code and thinking about binary numbers.
I don't see OS's really changing over the next 10 years even if dram does become more sram like and persistent over power outages, because all application code still would have to be architectured to use these features.
Come at me bros.
It'll just be like HP/Intel's Itanium
superior in every way, but ignored because nonfree shitware devs are too retarded to make their code base portable between archs.
oh jesus, stopped reading there
this thing is doomed before it even started.
>Its faster than everything else out there.
Companies do research projects all the time.
You can find papers about some of them on their website somewhere. Most of these are a long way off, if they're even practical and scalable.
When the company needs money, they say it's "just around the corner" to get people to invest. That doesn't mean it actually is.
It's the Kickstarter scam for big corporations.
>Thinking HP cares about the consumers with this
They're clearly targeting the high performance computing market. If their claims hold up this would be a holy grail in supercomputing.
It'll affect you if the cost of fabbing high density memristor dies ends up being cheaper than the cost of fabbing high density NAND like Samsung's 3D VNAND. Storage densities could skyrocket, and DRAM in all its forms could be a thing of the past though that paradigm shift is a lot less likely.
>look at how edgy and cynical I am
Assholes like you are a dime a dozen. Woefully ignorant on the subject at hand, yet all too willing to opine on it and detail how everything is really terrible but you're the only one who can see it.
This isn't a research paper. This isn't something that someone just pulled out of their asses to stop investors from pulling out. Memristors are well documented, they're a physically demonstrated and proven concept, the same goes for silicon photonics. Intel is HP's partner in codeveloping their optical interconnects for The Machine by the way. Unless you're insisting that intel is on this massive multinational scam, you might want to reconsider being such an edgelord. Its like I can hear your trenchcoat and fedora flapping in the wind right through my computer screen.
First tell me why you're so fucking stupid. If you really have to ask someone why a photon is better for signaling then an electron you're too stupid for this board, and that bar is already as low as could be.
So instead of using their ingenious memristors to make new memory, they're going to use it to make a full computer?
It's going to fail. They should've just stuck with making memristor SSDs. God dammit. Memristors will never take off now...
>he actually is this retarded
Do you have a learning disability or did you really drop out of school in the 5th grade?
Photons travel faster than electrons. The greater the distance that you have to carry a signal, the greater the benefit provided by using photons.
The less time data has to travel, the less latency is in the system.
Do you get it now, drop out? Do you need everything presented to you Barney style? Will bright colors and shapes make you feel better about being such a stupid little faggot?
>Photons are faster than electrons
i died a little inside
>using fucking Linux for a new OS
>using a harmful Kernel + Userland
>wanting GNU bloat
>wanting the toxic community and developers
Uriel was the hero we needed, but didn't deserve.
oh jesus fucking christ. Uriel was the same guy that blindly wholeheartedly supported golang (which has that oh-so-scary garbage collection that /g/ loves to hate for no real reason (none of you do any actual system-level programming anyway, you're all faggots)) and anything else that came out of rob pike's dick while calling any "new" or "difficult" concepts harmful (what he really meant by harmful was that "if an average person needs to use it, i can't use it because i'm a genius). he's no hero. he's a fucking retard, and he's better of in hell where self-convinced, self-appointed messiahs and self-important shitheads belong.
I found a leaked picture of the OS.
The UI looks very intuitive and polished, actually. I don't see how HP could screw this up - if they do this right, they could change the computing landscape forever...
Harold Finch was right.
There is nothing wrong with garbage collection.
If you had any significant experience in the software industry you'd see where Uriel & co.'s opinions quite justifiably originate.
Enjoy your bloated, buggy, unmaintainable, single-threaded, class-oriented code.
Don't do that to me, man. It took me a long time to come to terms with webOS' death. I don't ever want to go through that pain again.
>posting FUD that comes from armchair physicists on internet forums
HP's memristors are ReRAM. There is nothing uncertain about them. They exist. They're already in production. There is no issue with them violating any physical laws.
We had a thread several months back when they were talking about it in their conference and it made headlines.
I'm pretty sure this isn't a hoax or anything but they may be overselling some numbers especially for a first generation product they're going to sell. I will be more convinced when credible sources can cite numbers on its performance and such independent of the numbers HP provides.
I can't wait though until it makes it into the desktop computer, if it ever will, hopefully.
>oh my god, we've been using a similar computing architecture for 60 years
Stopped watching. This is as retarded as people who think we should switch to base 12 instead of base 10.
>huuur every base is base 10 face
It was an OS that was well beyond its time, especially when it first came out back in 2009. It was far beyond the level iOS and Android were at back then. webOS' multitasking and notification systems inspired the functionality seen in today's smartphones.
What, Linux++ will be removed and later they will use their own OS build from scratch called "Carbon". Linux++ will be used only so developers and similar ones can test their code and get familiar with "The Machine"