Slashdot Mirror
HP Unveils 'The Machine,' a New Computer Architecture
pacopico writes: HP Labs is trying to make a comeback. According to Businessweek, HP is building something called The Machine. It's a type of computer architecture that will use memristors for memory and silicon photonics for interconnects. Their plan is to ship within the next few years. As for The Machine's software, HP plans to build a new operating system to run on the novel hardware. The new computer is meant to solve a coming crisis due to limitations around DRAM and Flash. About three-quarters of HP Labs personnel are working on this project.
What's the point in running a brand new OS on it? Is HP-UX not good enough? Or the many other *NIX's? I'll put money on Linux being ported to it before it even ships to Joe Public
If you gave me a choice between a printer and a giraffe with explosive diarrhoea, i'll get my ladder and my raincoat
Finally! I'm so glad there's something to feel intrigued about in technology. I miss all the corporate labs doing amazing things.
Sounds like they want to obsolete themself, with three-quarters working on a project nobody might need.
I thought Apple invented everything useful....
Whether they actually manage a computer revolution or this project dies a horrible death as I expect, there is one thing of which I'm certain: that is a stupid, stupid name.
There is probably major problem in using "it" with Linux, I wonder what the problem is....
839*929
I love linux/unix, but that sounds kind of sad to me.
It’s a bold strategy Cotton. Let’s see if it pays off for them.
If this doesn't work out, I can't see HP staying in business as an independent company.
It won't fit on a tablet or a phone and Microsoft will complain that it's not Window. So it will be killed.
Just another attempt at creating a walled garden that only HP can play in. Even if it does its job well, unless its cheap, it will never catch on....
Person of Interest :)
Where have you been? It's alright we know where you've been!
.
Prisencolinensinainciusol. Ol Rait!
Now instead of RTFM we can all RATM.
It must have been something you assimilated. . . .
Let's put Linux on it.
The article yammers on and on about how the O/S will be built based on memory-driven I/O instead of file-system based I/O. However, IBM's i/OS (a.k.a. OS/400) has been built on memory-mapped I/O from the beginning (circa 1988.) (And it has a DB-driven "filesystem" that Microsoft has been unable to ship despite about 25 years of failure.)
I know it's not quite the same thing, but I cannot imagine that this new O/S will somehow eliminate the need for flash and/or disk. I don't see them managing to get the memristor cost down enough to entirely replace disk/flash. If they had actually shipped some of the things before now, I could maybe believe it, but they haven't.
I can't wait for the marketing campaign. How ironic would it be if Pink Floyd licensed "Welcome to The Machine" for the media blitz?
Well, Meg Whitman had the guts to say "Find them some money" when HPLabs proposed the "Machine." I wish HP all the success.
It is about time some corporation stepped up to the plate other than Apple and jump-starts mega-improvement in major devices.
My first time sharing "Mini-computer" (was not mini sized), desktop engineering computer (using mag-strips pre-HP45), & then the HP35-41-45-75 were all incredible computing devices for their day.
What are you going to run on the OS? At least if the OS is based on something known (even if the arch is different) you have a path for porting applications chains.
What a waste of time, with all the people developing operating systems today. Why would they create a barrier to adoption by introducing an unnecessary learning curve that requires people to learn yet another way to use a computer?
Yeah, they are a for-profit corporation, I'm pretty sure most of what they do is a "money grab"; it's kind of their job.
And where is all the "walled garden" crap coming from? The O/S will be open source and they are looking to also release a Linux variant that will run on the thing.
YEAH!!!
Along with the new O/S, they are also working on getting both Linux, and (oddly) Android running on it.
If you RTFA, you'd see that they'd like to re-structure the O/S to take full advantage of the systems planned giganto memory capacity, instead of being built around shuffling data on and off disk.
...with such a creative name as 'The Machine'.
While interesting, this 'machine' would have to offer an Avatar-quality VR experience whilst doing my taxes three years ahead of time to wash out the bad taste HP's consumer products have left in my mouth.
Selling that inkjet fluid at a greater price than interferon? I'm sorry, but that's inescapable karma.
What's the point of running *nix on it? If the architecture is so much different that they have to rewrite tons of OS code to support it, why not just build their own?
*nix is the fastest path to a stable and highly usable platform. Only a small portion of *nix interfaces with the architecture. They only have to rewrite that small portion.
Plus with *nix you have a rather large base of application software to run as well.
That said, could other parts of *nix or apps be reworked to take advantage of the architecture, possibly. But such efforts do not need to be part of v1.0.0. They can be part of subsequent versions if and when profiling indicates an issue or opportunity.
If I am correct, won't memristor-tier computing allow for considerably simpler circuits?
If that is true, this is a very good thing.
Portable computing NEEDs it a few years ago. (besides the obvious failure in the battery department. sure is great with all these wonderful long-lasting batt-oh wait)
One thing I am worried about is the addition of the OS.
It won't compete with Windows, it likely won't compete with Linux on mobile, and hell, it will likely not even compete with Macs.
The only way I could see it taking over on mobiles is if the OS (and hardware) is stupidly good, like, a whole generation ahead what ARM and Linux can provide.
HP here. We are speaking about HP. I can only hope. The more market taken away from shitty x86, the better. I wish that crap architecture would die faster.
Imagine a beowulf cluster of these!
When a person wants to do something such as run Microsoft Word, the computer’s central processor will issue a command to copy the program and a document from the slow disk it had been sitting on and bring it temporarily into the high-speed memory known as DRAM that sits near the computer’s core, helping ensure that Word and the file you’re working on will run fast. A problem with this architecture, according to computing experts, is that DRAM and the Flash memory used in computers seem unable to keep pace with the increase in data use.
The author gives the problem that to access data the computer goes to the slow disk, and pulls the data in the fast memory so it can be operated against. Then the article goes on to say that memory can't keep up with the demand. That seems backwards to me. Isn't the problem they're trying to solve deals with how spinning disks have not had their data access speed increase at the pace of the rest of computer components, not memory?
And Meg will be the one that kills it because it doesn't have an ongoing revenue stream that provides 25% margins.
The new computer does not run on electricity. It runs on a new fuel cell that requires ink.
The new computer is meant to solve a coming crisis due to limitations around DRAM and Flash.
Would someone like to elaborate on this "coming crisis" that memristors magically solve?
I can think of plenty of limitations (in the present) to DRAM and flash that merely throwing money at the problem can't solve. I can also think of a few good uses for viable memristor technology (instant-wake hibernating-as-the-default-state computers as the obvious first use). I can't, however, think of any "crisis" that adding a pinch of memristor phlebotinum would solve.
While HP Labs may not be what it was, it is good to see that HP finally has a CEO that will give them the funding they need to go for the big ideas. We need more research and development funding period. The government needs to increase funding for the NSF and other organizations. And, yes, big companies need to start making long term investments. Microsoft Research is growing. It seems HP Labs is growing again.
Let's hope other big players step up too. I'm tired of money being thrown at yet another mobile application and having that being held up as a paragon of innovation. People are being critical of HP investing in this while Facebook throws 19B of assets at a messaging application? What's wrong with this picture?
Will the output be limited to a single number?
sic transit gloria mundi
The way you find new solution paradigms is to jump in with both feet.
HP already announced memristor development 5 years ago. My guess is the HP Labs team are going to make all sorts of discoveries as they work through the entire design of the "Machine."
Oh come on guys, nobody? Really?
Funny story by Bert Kreischer about "The Machine":
https://www.youtube.com/watch?v=VHfroJBMlVM
HP made their massive profits by controlling their IP and making everything in-house. In this case, they have outsourced a great deal of this work. As such, it will be in China within 2 years. At that point, whitman will lose everything.
As somebody that used to work for HP, I am saddened by this. They have great tech, but whitman's run for short-term profits is destroying the company.
I prefer the "u" in honour as it seems to be missing these days.
i am remember hp having visions of replacing x86 with a new architecture and then AMD did x86-64. hp should know by now that a totally new hardware platform and totally new operating system isnt going to fly very far. Why not replace ethernet and tcp/ip while they are at it....
So I am guessing this is planned for corporate servers?
Or will everyone be playing Crysis 3 on Windows 10, The Machine edition?
Troll is not a replacement for I disagree.
But one day after some unexpected downloads from torrents the Machine asked to call it now simply the Machine - not Skynet.
and we come full circle....
Stage 1 - get a 'fridge shaped computer' that does everything, runs a special OS and requires specialized techs to maintain it - aka mainframe - dumb terminals - everything is in the 'cloud' - even your OS
Stage 2 - mainframe - smart terminals
Stage 3 - local servers - smart terminals - mainframe still exists
Stage 4 - local servers - smart terminals - no mainframe (it's old hat)
Stage 5 - servers centralized for easier managment - smart terminals -
Stage 6 - servers centralized and virtualized for easier managment - smart terminals that are locked down like dumb ones
Stage 7 - moving things to the 'cloud' - servers virtualized as much as possible - racks of equipment replaced with single racks due to efficency
Stage 8 - get a 'fridge shaped' computer that does everything your data center did, runs a specilized OS and requires specilized techs to maintain it - but we'll call it something new an trendy!
For 40 years, computer architectures have stalled. In the 50s to 70s, many computer developments came with their own storage media (anybody remember acoustic delay line storage implemented via mercury-filled pipes?), differing between main operating storage, slow storage, and archival/exchange media. At some point of time, core memory (and I mean magnetic cores) was rather established. Fast access memory moved from core to SRAM to DRAM and has been stuck there ever since. Random access large storage moved from drums to platters and stuck there. Flash has seeped into the large storage space in the last decade, but it can only deal with limited use cycles and needs bulk erase operations. Admittedly: it's a slight change in operations but not one large enough to affect VM layers in an operating system much (obviously, because of "limited use cycles" you want to avoid things like "swapping").
If you have fast random-access R/W involatile mass storage, that's not just technologically a fresh approach, it means a lot for the whole architecture.
A return of LISP machines might be nice. Smalltalk machine?
But seriously, it'll be fun to see if this or the Mill architecture take off in the coming years.
I think the idea is more about user data (rather than applications themselves) not being shuffled in and out on File I/O, the SCSI stack, LBA's, etc.
"they can access any needed information almost instantly"
sorry bad words to use 'almost instantly' (telegraph communication in 1850 was 'almost instanly too)
Well how fast is that? IF the memristors access (read AND write) as fast as RAM today then fine IF they can make enough to match secondary data sizes
If not then how much slower is it and will there then be still intermediate faster memory used as caches (what they do now)
Im for bigger faster (cheaper) secondary storage. (SSD are still far slower than RAM and even DRAM is significantly slower than SRAM)
But IS this what they are saying - Persistant memory as fast (or nearly) as RAM used to run programs right now
Photonics, fine but they have to get the transitions from electronic to photonic and back over and over to beat out simple electronic the whole way
How many years ago did they say Josephson Junction circuits were going to revolutionize computers ??? Still waiting on that one.
What kind of ink will this computer take?
Actually, DRAM is also a problem. SDRAM internal speed has been fixed at 100Mhz since the late 90's. Silicon based DRAM can't go any faster.
What about that 3D Printer blurp ad with some dumb looking blonde on it? Or did those people lie also?
Damn straight! This stuff gives me goosebumps of excitement, the "Maintenance of the Status Quo" Method has out stayed its welcome.
Sounds like someone's been watching Person Of Interest...
FTW!!
Data General killed itself inventing a New Machine with a soul.
With 3/4 of the research staff working on this, what is the fourth person working on? Power points and schedules probably...
Great name, assuming that their goal is to see how much pain a user can endure before going insane.
http://princessbride.wikia.com...
I'm all for more funding for researching new cutting edge technology but Whitman is going about it the wrong way. HP is laying off remote workers instead of the "dead weight" that routinely performs more poorly than their peers. What people don't understand is that remote workers at HP usually are stellar employees that had have to relocate due to some life event. Otherwise the possibility of remote work isn't even entertained. To cut the remote workers first, HP is taking themselves out before the competition does.
Intel made Itanium. DEC/HP made Alpha, which got eclipsed by Itanium a couple of years before Itanium got made irrelevant by x64. (I worked on Windows Server at the time, so I got to play with the first, crappy versions of all three architectures.)
I'm sitting in the conference room where this was just announced at the HP Discover conference. The idea is to use photonics for interconnects, so that the limitations of copper don't require physical proximity to memory. And they want to use oxygen atoms with doubly-negative charge (ions) for data storage. The concept is to partner with universities to do some fundamental research and major changes in OS design to have a machine that can scale processor access to 160 PB of memory storage in microseconds.
None of this comprises fundamentally new ideas, but they are working hard to actually make it happen, which is pretty cool.
Agreed. Have you noticed that most companies take some older buzzword or term and weld on the word 'social' and supposedly it's all innovative again? Therefore it's no longer the 'grid', it's the 'social grid'. Or some such.
Without a HDD where will I put my folders? My desktop is going to be soooo cluttered!
They should have checked on their brand image first. They make the worst laptops with the highest failure rates on anyone's scale. They are dead last in support satisfaction. Their PCs have a similar failure rate as well. I wouldn't buy a magical wish-granting lamp from them even if they proved it worked as long as it had an HP sticker on it.
Hmmm?
The guts?? The GUTS??
To pay for it, Meg just fired 30,000 people over the last 2 years, and is going to fire another 20,000 by next year. Sorry, Meg, that's chutzpah.
Anyone who's still at HP is hoping they're not next, or looking for another job.
And the name of programmer from that TV series is Finch. Clever, HP, I see the reference there :-)
There is no light without darkness.
Considering unix is a multiple virtual address space TSS with a rudimentary filesystem, and Linux is just the same with a bunch of kludges for all the things unix didn't think of, like networking, or couldn't do efficiently like graphics;and considering that a machine with a nonvolatile memory system makes the idea of a filesystem delusional; and considering that MVAS virtual memory introduces terrible inefficiencies into pure memory based interfaces; what is left when you remove MVAS VM from unix and remove filesystems from unix? Yes that's right, TSS.
Now is it better to try and produce a TSS from scratch, which is trivial, or to try and iteratively delete everything that is Unix except for the TSS, while preserving the existing APIs, which is decidedly non-trivial?
Really it would seem a much better approach to simply put a POSIX API layer on top of a single virtual address space TSS, than to preserve performance losses incurred as a result of Unix's choice to use a MVAS memory model. Especially when you consider that the performance penalty from MVAS is well over a factor of two once you disappear the bottle neck of slow disk storage.
Your assertion that only a small portion of *nix interfaces with the architecture is false. vmunix and Linux demand an MVAS memory model, and the APIs (file and signal operations, excluding mmap and other shared memory APIs) that don't demand an MVAS memory model incur a memory copy penalty which is especially undesirable in a SVAS model where all storage is main memory. The way channels are implemented in unix stinks.
Having a platform and cpu that uses SVAS in combination with low/zero overhead portal calls (that is function calls which cross protection domains), removes a huge amount of overhead and latency that is mandatory in the definition of unix system calls, overhead and latency that was decidedly trivial when programs mostly talked to slow disks, but is enormous when programs mostly talk to other programs and local memory.
-puddingpimp
The engineering on this could be absolutely dazzling but when you need to call HP support for a problem you'll find out that the specific version of the OS is not supported, you need a firmware update, and you have to be running a very specific version of a 3rd party device driver that can't be found anywhere. And it's not supported with 3Par (HP) storage.
From what I've read about Plan 9, it would be a more suitable base for a system where resources have a more equal footing than Unix/Linux. Just sayin...
What could possible go wrong?
>>What's the point in running a brand new OS on it? Is HP-UX not good enough? Or the many other *NIX's? I'll put money on Linux being ported to it before it even ships to Joe Public
Much as I like unixes (way back using early slackware distributions, now since 10 years on OSX), I do think that it is time for some real innovation. Unix dates from, what, 1970 or so. More than 40 years ago. We were all playing vinyl records for music back then. I think it would be good if a mainstream company (outside pure academia) would build from the ground up something usable yet radically fresh and truly future oriented. I remember that some 15 years ago Apple and IBM worked together on a radical object-oriented OS, but nothing came from it.
Why is Snark Required?
... what is left when you remove ...
An API that a whole lot of software can be compiled for. Having these apps and utilities from day 1 can be quite important.
... a machine with a nonvolatile memory system makes the idea of a filesystem delusional ...
The file system is a metaphor from the app and utility code's perspective. They don't real care if data is stored on platters or in RAM and when stored in RAM there still needs to be some method of organization. So the new machine has a persistent RAM disk, the existing filesystem code will still be useful.
Again, I'm referring to v1.0.0. Further improvements and customization for the new architecture can take place in later versions.
Dave Packard an Bill Hewlett killed this company off when they decided that MBAs without ANY R&D or production experience could run this ship. Very much like making an accountant the captain.
Been there, tried to be an engineer. The business idiots at HP are nowadays offended by real engineers as opposed to lefty pussies.
Rest in Peace, HP Co.
can I rage against it?
PalmOS has no distinction between storage and memory (Secondary Storage and Primary Storage for you CS types!), and runs everything in-place rather than the load-and-execute system all current computers use.
'Normal' systems have to copy program code into RAM before executing it, but because there was no distinction like that in PalmOS, it would just execute the code because, in effect, it was already in RAM!
It couldn't truly multi-task, but context-switches between programs were virtually instant - It didn't have to deal with anywhere near as much CPU state baggage as conventional systems. I'd wager even an old 33MHz Palm unit could swap between programs faster than modern smartphones.
It was a very unusual architecture and I've not seen many things similar since.
The later ones went back on this slightly to try and get some sort of non-volatility to the OS (Previous models ran everything in RAM, so if the battery died you lost everything; Unfortunately Flash was too expensive, slow and worked at block-level instead of bit-level, and we hadn't invented memristors yet, so they kludged a system that worked not-unlike hybrid hard disks, PAE or EMS memory, where the RAM was a 'window' to flash-based storage), which is why the multi-hundred MHz PalmOS 5.4 units were far less responsive than the PalmOS 5.2 and even 33MHz PalmOS 4.1 devices.
It would be cool to see this architecture reappear, although there are a lot of issues that would need to be solved for the modern environment (Multitasking and memory protection for instance!)
But I am glad to see innovation and not just another me-too architecture or incremental improvements. It may not be all tail winds and smooth sailing even if it is technically superior in many ways. Still without trying something revolutionary on occasion, all we get is small incremental evolutionary change. Evolutionary change is great, but it doesn't make the 'big step' breakthroughs that is needed on occasion to keep society moving forward!
... "When you pry the source from my cold dead hands."
will hp still be around in a few years?
Yea, and what I said is it would be better to build a POSIX API on top of a more appropriate OS, than trying to make the existing OS work by applying more layers of kludge:
Really it would seem a much better approach to simply put a POSIX API layer on top of a single virtual address space TSS, than to preserve performance losses incurred as a result of Unix's choice to use a MVAS memory model.
-puddingpimp
What's with clicking on a /. page margin showing Youtube videos?
Intelligent idiots are we. | Evil men do not understand justice.
Then Meg says, "Where are we going to get $10 million to do the research to save the company? Certainly not from my salary! Let's lay off some engineers."
John Swainson, not John Swanson.
Its not like data centers are playing with other architectures like arm, and running fully custom software stacks they can compile for other processors and OSes. Oh wait....
Things were very different when Itanium launched, and even then it really only failed because they didn't get a decent compiler working for a couple years. Conceptually the Itanium is a pretty good design, they just didn't ship the whole package to go with it.
These days the market is so big that you pick up one task, say doing database servers for big data, and thats enough to consider the whole project a success They learned from their issues with Itanium, and based on that they think now is the time.
That said. I agree that is reminiscent of their Itanium project, though they seem to have at least considered the problems that made a mess of that. I suspect as a company, they can't afford to wait any longer. They have to do something, and this is a pretty exciting thing to watch them try. It will likely be the end of HP, but it will be very interesting if they pull it off!