Slashdot Mirror
Intel Expands Core Concept for Chips
Aziabel writes "As most of you have probably heard, Intel plans to come out with chips containing two processing cores next year, but that's just the start. The Santa Clara, Calif.-based chip giant intends to exploit the concept of using multiple processor cores; chips with four cores and eight cores will eventually join dual-core chips, which will begin to appear from Intel next year. The company's research department is also looking at the feasibility of creating chips with hundreds of cores to assist servers and supercomputers with large numbers of relatively repetitive calculations, said Steve Smith, vice president of the desktop platforms group at Intel. The focus on multiple cores arises from Moore's Law, which dictates that the number of transistors on a chip doubles every two years. I say, the more the better. Keep 'em coming, chip-makers!"
...a beowulf cluster on a chip!
How am I supposed to fit a pithy, relevant quote into 120 characters?
Becha can't eat just one!
It's nothing more than a catch-up move to Sony/Toshiba/IBM Cell, just like EMT64 to catch up AMD. Those late and awkward moves are of bad omen for Intel, IMO.
The thought of playing Battlefield on a dual core Opteron. Actually, just having a dual core Opteron for any use has a serious drool factor.
Professional Politicians are not the solution, they ARE the problem.
I am beginning to suspect that Intel does things like this simply to make x86's instruction set harder and harder to emulate well.
Kind of like to what I suspect Microsoft has been trying to do against Lindows for a while now, namely complicate their API more and more. And with IE and HTML.
Of course they're well within their rights to try. We'll just build a better idiot savant. Or let Steve Jobs keep making Apples that no one can really imitate in the first place.
vicious, untreated political sewage...niche entertainment for the spiritually unattractive...worshipless pap
"The focus on multiple cores arises from Moore's Law, which dictates that the number of transistors on a chip doubles every two years. I say, the more the better. Keep 'em coming, chip-makers!"
No. I think it arises from the limits of the von neuman architecture.
This does not bode well for problems that mathmatically cannot be executed in parallel.
Now I can do away with my furnace.
The problem is with what they (both intel and amd) plan to do is saying a dual core 1.5 centrino (for example) cpu is actually a 3Ghz machine (from the pr they have allready put out about these chips).
Read overclockers.com for some good speculation on what the good/bad/ugly features are likely to be.
...
The focus on multiple cores arises from Moore's Law, which dictates that the number of transistors on a chip doubles every two years.
I don't think non-compliance with Moore's Law is a felony. It's an observation, not a statute. Moore's Law arises from the fact that transistor counts keep doubling, not the other way around.
Also, doubling the number of transistors in any way possible doesn't necessarily translate into double the power for any given application. In this case, multiple cores are good news for multi-threaded or forking server apps, but rather less interesting for a lot of desktop apps. Intel obviously has a vested interest in pushing ever larger die sizes, because it does large dies better than anyone else. Whether this will always be in the interests of the rest of the industry, let alone the end user, is less obvious.
Virtually serving coffee
How useful is this going to be when a alrge number of our apps (games, games!!) are written with a single processor in mind? seems to me like a case of gaming on a dual-xeon/operton. Waste of money and resources.
if a kernel is written to take advantage of multiple cores, would this mean applications written ontop of it would start using the multiple cores?
if not, how feasable is a multicore > single core emulation in linux.
>will begin to appear from Intel next year.
Very likely this is marketing sp33k for "will be paper-launched at the last day of next year"
Belief is the currency of delusion.
Supercomputers my ass, when will Intel admit that the real driving force behind faster hardware is lots of gibbing and blood splashing around the screen in real-time! This sounds like a very good idea and one that could maybe lead to a demise in separate graphics cards? your graphics could be handled on a separate core that gets its instructions from another core that maybe even separates collision detection and AI into other cores? I know purpose built hardware is always going to be faster at its job, but if this becomes much cheaper, it will just be more economical to ditch your graphics card and get more cores?
This comment does not represent the views or opinions of the user.
imagine how much power is used and heat generated by 2,4,8,16,....256 cores. watch your electric bill go way up. toast anyone?
Head of Intel: Today we announce our new 8 core processor, that's a nice addition to our 2 and 4 core processors released earlier this year.
Officer of Law: you are under arrest for breaking the Moores Law, that allows you only double the number of transistors within a year.
IBM have been doing this for years - and its biggest technological success story, the POWER5 chip shows that Intel are blatantly only playing catch-up with this announcement.
5 _moores_law/ shows this perfectly.
http://www.theregister.co.uk/2004/11/26/ibm_power
It reminds me of the way that Intel pretended that they invented integrated wireless technology with its Centrino chip only after Apple had been shipping laptops for nearly two years with internal wireless cards.
Normally, asking if they had no shame would be appropriate but it is unfortunately clear (without the need to ask) that they don't.
Most of the reports that I have read have said that AMD will be releasing theirs next year and Intel the following year. Intel, though didn't start talking of dual cores until AMD started talking about theirs. From research that I have done, each manufactorer has some mighty issues to overcome with the single core before dual cores can be implemented nicely.
AMD has said that dual cores will be clocked anywhere from 600Mhz to 1Ghz slower than the single core counterpart, namely because of heat issues. There are many more issues that arise with dual cores here are a few
Cache correnance
Bus contention
software implementation
plus more
It will be interesting none the less on how each manufactorer overcomes the issues with multi-core chips and the benefits to the user of of multi-core.
Just wondering.
A few years ago I thought of a different kind of twist on computer architecture that I labelled OOH.
The basic idea is that a computer could comprise many, many tiny CPUs, each with its own tiny local memory.
A given (CPU+RAM) could be designated to operate as RAM for another CPU, so the MMU/OS could balance the number of processes needing memory with those needing processors.
A (CPU+RAM) could also be labeled as a slave to others, so a multithreaded application could have the number of processors it needed.
I haven't thought about it in a while, and it's been some time since I studied architecture, so probably these ideas are hopelessly naive.
Raise your children as if you were teaching them to raise your grandchildren, because you are.
Generally, bash is superior to python in those environments where python is not installed.
Years ago (early 90s?) there was a lot of talk about "wafer computing", but it never seemed to come to anything. Maybe now we'll see it take off.
When manufactufing chips, they're done so in wafers. Then the wafer is cut up into its component parts, and each part is sealed in its own case. It would seem to be more efficient to just stick the whole damn wafer in a single case.
It would give a whole new meaning to "pizza box server", as the wafer and case would closely match the size of a pizza and box, respectively.
According to Sun CEO Jonathan Schwartz they have 8 way chips already.
from the rated-T-for-thank-god-thats-over dept.
Something like Linux SMP will see each core as a separate CPU and treat them as such, much as it does with hyperthreading today. The catch is, your applications have to be multithreaded. Then they will take advantage of the multiple cores.
How am I supposed to fit a pithy, relevant quote into 120 characters?
Sun's new Ultrasparc IV shipped in the SunFire 490's and larger servers already do this. The plans right now are to scale this up to 32 cores per cpu. The only issue that I see is that the memory controller is onboard the cpu, so while you may have 2/4/8/16/32 cores, you still only have a single memory controller, which limits the ammount of ram you can have. I'm sure they have a solution for this, but I don't know what it is.
This signature is a waste of 42 characters
Yeah, it is called Niagara, and it is working silicon now, but far from done. expect an unveiling in February.
If you want to know a bit more about it, I wrote it up a few weeks ago here:
http://www.theinquirer.net/?article=19423
-Charlie
WTF?
Laws are for people with no friends.
Intel just canned their 8-way chip and replaced it with a variant of Montecito, or more likely a Montvale derivative. Here is a bit on it:t tp://www.theinquirer.net/?article=20286
:)
http://www.theinquirer.net/?article=20270
h
Needless to say, their long term strategies are a tad up in the air right now.
As for their desktop (IE P4 based) dual core plans, there are 2 generations planned. The first is a simple pairing of 2 current cores with a minimum of tweaks, basically a scared response to AMD. The second one is really the first one they planned, and it is a lot more sophisticated.
AMD was there from long before Day One, and have the most coherent philosophy on dual cores for the desktop/server.
Rather than re-write all my own articles here, here is a link where I break down all of Intel's dual core plans as well as some of AMDs.
http://www.theinquirer.net/?article=17906
Sorry for all the self links, but I don't really want to keep re-writing that stuff, links are the reason behind the web, right?
-Charlie
Will this hardware development bring quantum computing any closer to reality?
Although for some, non-memory intensive, highly threaded applications multiple cores can be a boon, for many applications this won't be a boost in performance at all.
Remember that each of these processing cores will have to share their memory bandwidth and possibly level 2 cache as well. As it is Intel's EM64T Xeon processors really feel the bandwidth bottleneck in their memory interface and can easily saturate it.
I can see a dual core Xeon being able to saturate its memory bus on its own. Similarly, the dual core Opteron, unlike a dual processor Opteron, will have to share a single memory bus and hence be slower than a dual processor machine.
Adding extra cores merely moves the computing bottleneck elsewhere, it's not a panacea.
Agrajag: "Oh no, not again!"
"Laws" like Moore's, Newton's, Ohm's and others, don't "dictate" anything. They "describe" observations. Intel doesn't meet integration targets based on some hoary old directive from Gordon Moore from the late 1960s. They meet production deadlines projected as close to their maximum productivity. Moore observed the logarithmic rate of transistor integration increase way back then, and described it as invariable as gravity.
Engineers especially must understand that "laws" of nature, including human innovation, are governed by an "invisible hand". Not some imaginary deity, or some government, or some mythic genius. Rather, there is a deeper order to events, like the way every triangle has 180 degrees, the Sun "comes up" every morning, controversial Slashdot posts will get mod'ded "Troll", without any false statements or duplicity. We're engineers: our job is to engage the deeper order, understand it, model it, and exploit it, without further mystifying it.
--
make install -not war
I know some are pointing to the Cell project as the inspiration here, but Tera was hard at work on this long ago in the form of the MTA
The MTA was a commercial failure. Tera's inability to execute as a company was a major reason.
It is fun to watch Intel chase AMD.
Intel has been roadmapping this sort of thing for over 10 years. They got distracted by whoring to the Internet and AMD's 64-bit overreach.
Besides, putting a math coprocessor alongside every integer unit was the beginning of multi-core CPUs.
Quantum computing will come to the rescue to many of the math problems of which you write. For many types of problems that are NP-Complete (i.e. are like the travelling-salesman problem) it will reduce the need for multi-core processors. Of course the more processing power the better!
Given that I can't get good working 3d drivers for FreeBSD (by which I mean I can upgrade my kernel from time to time without worry), or Linux; on any fast 3d video card, I'm looking for anything that will give me good 3d graphics without hasstle.
Some big ass 200mm wafer for a processor. Like, they trim the crust off, slap a heat spreader on it, and send it to the mainboard manufacturer.
That would be, like, totally awesome. You could play Doom3 and Half Life 2 at high quality.
However, you could also convert the screener of "Doom" into xvid/mp3 within a few minutes. Which means it's available faster, which means there will be more niave bastards downloading the thing. Which means more psychiatric evaulations, and more medicine.
*invests in Pfiser, GlaxoSmithKline, and Merck*
"The focus on multiple cores arises from Moore's Law, which dictates that the number of transistors on a chip doubles every two years."
yeah its because of moore's law. you are such a frigging idiot.
Why stick up for big business?
Intel is, no make that was, rumored to be, [no, definitely are] in the process of buying the design group that develops Itanium from HP.
The vnunet page has a little speculation as to why the move is being made. But if you put that together with HP's general strategy of streamlining its fragmented high performance server offerings: Then the picture that emerges is in agreement with parent comment: Intel is in catch-up mode. They have, as other stories and commenters have pointed out in
SLASHDOT: news for people who can't concentrate on work or have no life at all and got tired of yelling back at the TV.
I'm sorry, but I've had it up to here with cowboy neal's commentary. While this intel thing with the polycore procs is a nifty happening, what it really is is intel catching up with IBM and motorola.
Really, neal. The G series has had this architecture for years, and you've never once mentioned it, implying that intel is some sort of haven for intention. Anyone who's been in this business more than 15 years knows they're not...they've ALWAYS played catch-up with other chip manufacturers.
Why wait? AMD has this now and it appears Intel is now following AMD in this direction.
AMD Press Release
Because it's such a leap from dual-core (which everyone has been thinking about for years) to multi-core.
I mean, they _had_ to be copying Cell, right? Otherwise, they would have had to hire a bunch of ex-IBM'ers or Sony people to help them think of scaling even higher than 2.
Ridiculous.
This is much less of a problem than you might thin, not because it isn't a real problem, but because it is so obvious. Everyone already has a workaround, most of which involve FB-DIMMs.
Niagara (see my post above) is bandwidth rich, the AMD solutions are also. The only ones with a looming problem are Intel until CSI comes on in a few years, but that is manageable.
Moral, Sun OK, AMD OK, Intel solid plan.
-Charlie
Shouldn't that be, the Moore, the better?
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Some software is licensed "per CPU" - the more CPUs you have, the more you pay.
But what exactly IS a CPU from a software-licensing perspective?
I anticipate a few court cases over this, particularly involving small- and medium-sized software vendors and anyone selling high-dollar software.
Big players selling sub-$500/cpu software will hopefully be more interested in goodwill than greed and will offer the market reasonable pricing for these not-quite-multi-cpu computers.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
period porn?
What you're saying is, for applications with poor cache performance, multi-core processors will be no better than single-core. Personally, I can live with that. Most of the processor developments of the last 10 years have favored applications with good cache performance.
What worries me is what happens when the OS schedules a process with good cache performance on one core, and a process with poor cache performance on the other core. Unless the cache does something special to prevent it, the "bad" process will completely deplete the cache, causing the "good" process to slow way down.
I recently worked on a real-time program for the Pentium IV, and we found that our worst-case performance was actually 4-5 times worse when hyperthreading was enabled, because our process would occasionally have to share its cache with something that was heavily memory-bound.
Shouldn't you be surrendering to someone?
It's because they can't speed up the clock rate any more. Nobody wants to admit it so they switch over to multicores and try to distract you from the fact that it's the same clock as in your current computer. They're terrified that people are going to stop upgrading.
What!??
p00??
I can't disagree.
I think at that point it would be the mainboard.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
The ultimate limit on the number of cores you can put on a single chip is the available pin bandwidth. At a point, there simply isn't enough bandwidth available to supply instructions and data.
Sun took this further with Niagra and got rid of the pipeline and OoO which gives them a lot more room to fit more cores on the chip. So threads on Niagra will have a lot more memory latency but you will be able to execute a lot more threads.
The Cell processor is nothing but hype right now. The main concern here is what kind of DRM they are implementing in Cell. Cell's on-chip DRM should take highest precedent instead of its technological leaps. If anyone can subdue their awe for Cell's unproven performance claims, please try to remember the Pentium III's Processor Serial Number issue. Here's a reminder:
9 /GroupG/psn_outline.html
http://www.sims.berkeley.edu/
courses/is224/s9
Sony and Toshiba have their fair share of interest in the entertainment industry. Toshiba have shares or own EMI, one of the big music label. It's early to speculate, but I'll bet their goals are to implement DRM on a global scale. The PS3 is their ticket, then they'll move to computers and electronic devices.
can it play Duke Nukem Atomic Edition?
The higher the technology, the sharper that two-edged sword.
Even with quantum computing, you are eventually going to run into a brick wall where you simply cannot compute things faster. From there you can try to put things in parallel, but you can only do that for so long. I guess if you could go faster than the speed of light and squeeze a computer into a singularity then there might be no limit...
Or you could do your computing in space traveling near the speed of light...
http://en.wikipedia.org/wiki/Transputer/
We already do this; integrated graphics use a 'seperate core', indeed they use a seperate chip. What would be the advantage if we put these integrated graphics cores with the cpu? In part, seperate cards are fast because they have dedicated RAM that's always one generation ahead of system RAM. Somehow this is the currently preferred choice even though that dedicated RAM is wasted when not engaging in gibberific blood-splashery. If the current market doesn't sell good 'graphics' MoBos, why would Intel/AMD bundle graphics in their cpus?
Great. Just what we need. There are a lot of programmers that will try to take advantage of multiple cores, but very few programmers are able to write properly-threadsafe code. Bring on the bugs.
I don't know much abut AMD roadmap, but I can guarantee you that intels roadmap for 2006-2007 is gonna kick ass....
the race for Dual core will be won by AMD by a short time. The next 2-3 years will be hard for intel, but after that time both companies will be tied in a neck to neck race.
intel advantage is the volume, resources and incredible manufacturing capacity.
AMD is the better product they have now,and the momemtum they are gaining.
Intels manufacturing capacity means they can produce faster/smaller/cooler transistors than anyone else.. by far (i've seen the charts) I mean, intels 90nm is better (powerwise) than TSMC 65nm will be!!!.
Let's see... Wolfenstein 3D traced one ray for each of 320 columns of pixels in real time on a 25 MHz 386 machine. Assuming this window is 200 pixels tall (leaving 40 pixels for a statusbar), one could do "true" raytracing on a hypothetical 5 GHz 386, and with the better instructions-per-clock of a modern CPU, we're looking at 640x400 pixel raytracing in the very near future.
To include "Moore's Law", which is an educated (and heretofore correct), guess at the progress of transistor design, with laws that consitently and correctly dictate how the world around us works, is silly.
Q: What do you think about American Culture?
A: I think it's a good idea.
(adapted from Gandhi)
it is difficult (sometimes impossible) to extract parallelism from the majority of the desktop computing applications
You mean it's difficult for the compiler to extract parallelism from pure sequential code. However, a smart programmer could easily make all sorts of tasks run in a separate thread from the main UI thread:
You say multithreaded programming is hard? Tough cookies. Early adopters of desktop SMP will demand multithreaded apps that take advantage of the whole computer. Besides, a "revolutionary SDK that allows easy multithreading" would probably just be language facilities for better threading and/or a thread debugger.
And that's not to mention background apps that could run without impacting the main CPU:
In short, I see the point of desktop SMP as to look at the big picture, to try to extract parallelism from a single thread but to extract obvious opportunities for splitting a user experience into multiple threads where possible.
I don't think non-compliance with Moore's Law is a felony.
Competition enforces it. If 18 months pass, and you don't have a smaller process, then your competitors will fine you by taking your customers even if some government doesn't. That's how Mr. Moore's observation becomes law.
multiple cores are good news for multi-threaded or forking server apps, but rather less interesting for a lot of desktop apps.
Desktop apps can be multi-threaded, and most users have more than one app running at once. Read More...
claiming to invent a technology? They aren't doing something new, but it is new to the consumer market.
So multi-processor OS is nothing new.
It is if it's Windows XP Home Edition. Some app vendors have already stated that they will charge more for apps designed for a multicore CPU.
Apple only made it asthetically pleasing by putting the same damn card "in" the notebook instead of sticking out. thats all. Centrino the chip is on the motherboard like a normal network card. Placing an embedded chip vs. just taking the clutter out of the way by placing an "internal card" is very different
A given (CPU+RAM) could be designated to operate as RAM for another CPU, so the MMU/OS could balance the number of processes needing memory with those needing processors.
Your "OOH" (object oriented hardware?) design is technically called NUMA. An operating system could naively implement NUMA by, say, storing a CPU's swap file on a RAM disk spread around multiple CPUs, although real NUMA architectures implement something more sophisticated.
laws that consitently and correctly dictate how the world around us works
Physical laws such as Newton's and Ohms are also "educated (and heretofore correct), g'uess[es] at" the behavior of matter and energy, observations which are made by fallible man. Newton's laws fail at submicroscopic distances and additionally at high velocities. Ohm's law fails on non-ohmic materials. Likewise, Moore's law may fail at specific points on the timeline of production.
You are sort of right on this, I know what both companies are up to. Merom/Conroe/Whitefield will be fine chips, as will K10. I think AMD will beat Intel out of the gate with the next gen cores though.
As forthe processes, read the Inq for more, AMD/IBM have come a long long way recently.
-Charlie
But what exactly IS a CPU from a software-licensing perspective?
Why the hell would a software license care about the number of active CPUs on a board?
Of course, adding multiple cores is just the easy way out for computer architects. There's nothing magical about putting multiple cores on the same chip. It just increases overhead. A single core with twice transistors can be more powerful than two cores as long as the single core is well-designed.
With two cores, pretty much everything is duplicated, so some work must inevitably be duplicated as well. The transistors could be better spent expanding the L1 cache and widening the hyperthreaded superscalar pipeline.
It seems to me like you could parallelize most desktop apps.
An MP3 player: I would think you should be able to decode one second with one processor, and the next second with the other processor.
Word processor: I would think that parts of the boot process that do not require the other parts would be able to run independently. Two processors could check alternating lines until the whole document was checked.
Spreadsheets: I would think that the first half of a giant list could be handled by one processor, the second half of the giant list handled by the second processor.
3D graphics, non-accelerated: I would think that the screen could be cut in four, and one processor rendering each part.
Games: I would think the simulation could be divided into parts, and the different parts simulated by different processors.
Compiling, Parsing: If you have 40 files to parse, each processor can handle 10 files, so you would go roughly 4x as fast. Parsers aren't just for C++, parsers are found in just about every program that reads data from a disk or the Internet.
So it seems to me that we could make major performance gains, using multiple processors.
Yes, that's right. But it's clustering that scales for anything from a coke machine to a supercomputer, working under the same model at the board level as on a worldwide cluster of machines.
Or it would, if it existed.
Raise your children as if you were teaching them to raise your grandchildren, because you are.
I read that some functional programming languages can automatically multithread a program so that the task is split up over multiple processors. The programmer would just program as for a single CPU and change nothing or very little.
/. thread:
Functional programmming languages examples are Lisp and OCaml.
Oh, correction, from a previous
OTOH, it is theoretically possible to automatically multithread purely functional programs, especially if they're lazy like Haskell. So it could end up being a very important language on multi-processor and distributed systems.
The only way I see multi-core processors or cluster-like processors (Cell) succeed is if programmers switch to languages like that. Any other way would introduce too many bugs in programs. Computers should make life easier, not harder. Even for programmers.
Eventually, multi-core/processor is the only way forward, long before single-processors have to heat up to supernova temperatures to increase speed.
We're just at the beginning of computing. Looking back, programmers of the future will pity us poor folk who had to make do with only 1 CPU. However, we need the right tools to move forward. Anyone know if there's an automatically multithreading (functional) programming language in existance or being invented?
- -- Truth addict for life.
the x86 instruction set is.
intel either implements it or emulates it in order to run all that stuff that is.
if intel makes a new instruction, it's FOR programmers, not something to be hidden from them.
links are the reason behind the web, right?
Then at least html-ize the links!
http://www.theinquirer.net/?article=20270
http://www.theinquirer.net/?article=20286
http://www.theinquirer.net/?article=17906
-d
Love many, trust a few, do harm to none.
http://shit.slashdot.org/article.pl?sid=04/12/18/1 413213
Threading is a design level optimization. A programmer won't generate "pure sequential code" if the design incorporates threads in a clean way. No, I don't know whether university professors teach threaded programming to their students. What has been everyone else's experience w.r.t. this?
Though the momentum of their girth will carry them a bit further they have been too slow to move.
"If any question why we died, Tell them because our fathers lied."
Been there, doing that.
It annoys me how magazines as well as Intel tries to portray ideas that are already being developed or developed in the market like it was they who came up with the new innovative idea.
Dual Cores, sorry, buddy but IBM and AMD have the lead on this one. Multi-Cores, IBM and AMD's design already takes into account multi coring, not only that by the Hyper Transport concept for AMD specifically had multi-coring in mind when it was introduced more than 3 years ago.
Intel is a great company, but, I wish magazine editors would stop making announcements like all of sudden Intel has done something new every time Intel's marketing team goes to pitch.