Slashdot Mirror
Intel Launches 72-Core Knight's Landing Xeon Phi Supercomputer Chip (hothardware.com)
MojoKid writes: Intel announced a new version of their Xeon Phi line-up today, otherwise known as Knight's Landing. Whatever you want to call it, the pre-production chip is a 72-core coprocessor solution manufactured on a 14nm process with 3D Tri-Gate transistors. The family of coprocessors is built around Intel's MIC (Many Integrated Core) architecture which itself is part of a larger PCI-E add-in card solution for supercomputing applications. Knight's Landing succeeds the current version of Xeon Phi, codenamed Knight's Corner, which has up to 61 cores. The new Knight's Landing chip ups the ante with double-precision performance exceeding 3 teraflops and over 8 teraflops of single-precision performance. It also has 16GB of on-package MCDRAM memory, which Intel says is five times more power efficient as GDDR5 and three times as dense.
"The cores are 14nm versions of Silvermont, rather than 22nm P54C"
So, somewhere someone at AMD is going "fuck it, we're going to 128 cores".
Damn ... that's a crap pile of cores ... that's like, Skynet in a box or something.
The mind reels.
Lost at C:>. Found at C.
I want one of these in my next notebook....
Just saying
It is probably a good chip for it's niche, so you would think they would have less bloviation in their intro video. If this was anyone else I would assume they were mostly trying to fleece more investors before they inevitably went belly up. It's so bad that major league sports style animation with yelling pitchman and a pounding beat would be an improvement. That bad.
Why is Snark Required?
So what exactly is the real world application of such a beast? Are there that many x64 based supercomputers out there?
I also wonder the reason behind Intel's decision on using 22nm on the Knightslanding instead of the latest sub-10mn node that it has
Perhaps the latest node is not stable - or perhaps Intel wants to tap out the max value of whatever they had invested in the old 22nm node
So Intel is top dog given that nVidia is only producing 2.3Tflops, right?
I guess AMD gave up on HPC. If I read the wiki right, their top card does 0.1Tflops
McRAM?
Yes, I would fries with that.
I've been asleep for 20 years so I guess CISC won?
Zoid.com
Windows response, a big red letter message:
"WTF?"
Religion: The greatest weapon of mass destruction of all time
...but I suppose 640 kilocores should be enough for anybody.
So how fast can it calculate a Knight's Tour
I eat only the real part of complex carbohydrates.
Defects in the process bleeding edge process are the main reason to use the older process. When they make one of these insane multi-core parts the die size is very large (sometimes taking up a whole 26 by 32mm scanner field) thus the yields are hit harder by defects. On a more consumer level chip they may have 4 or more die in a scanner field. A single defect in this field will take out one of the four die resulting a a yield of 75% for that field. However in the case of a single die for the whole field the yield would be zero with the exact same number of defects per mm^2. I am sure they have a greater understanding of where their defects come from on the older 22nm process these days and can ensure good yields even with a huge die size.
An additional reason they would use the older process is a chip of this level of complexity probably requires tighter overlay and critical dimension (CD) control than the "standard" 22nm process to work well. Having a well defined process makes tuning all of these factors much easier and it also helps decouple if it was it the process or possibly a issue in the design when initial silicon runs do not work exactly as intended.
This has been corrected in the post. Intel is in fact using their 14nm node for Knights Landing.
Or be able to load Windows 11?
You are truly sick. Get some help. You, and the one who modded you insightful.
All I did was provide a useful link. And you go nuclear-fractal about it, exploding with invective and unsubstantiated speculation.
If it weren't for deadlines, nothing would be late.
I'm pretty sure that Anonymous Coward's response was itself propagating some other kind of meme, but I'm not motivated enough to look it up.
Anyway, the thing is to take inappropriately over-the-top invective with a grain of salt, since it was probably intended to be tongue-in-cheek.
I don't care if it's 90,000 hectares. That lake was not my doing.
Are you a well adjusted, happy person? Ima guessing no.
Excellent! Now we'll be able to process even more bullshit widgets on websites!
I am still annoyed that Skylake still only comes with 4 meager cores and some lousy graphics I will mever make use of, and anything beyond that is a hockey stick price increase. Taunting us with 72 is just cruel.
While supercomputing is a very small section of the computing world, it's not that hard to understand.
First of all, this would make for a terrible graphics card. This (deliberately) sits between a CPU and GPU. Each core in a Phi has more branching support, memory space, more complex instructions, etc than a GPU core, but is still more limited than a Xeon core (but it has wider SIMD paths).
A GPU has many more cores that have a much more limited set of operations, which is what is needed for rapid graphics render. But, those limited sets of operations can also very useful in scientific computing.
I haven't seen anybody try a three pronged approach (CPU/Phi/Nvidia Tesla), but I will admit I didn't look very hard. This is all in the name of solving really big problems.
I have eight (8) cores on my laptop. Frequently, a single multiprocessor-unaware application will hog an entire core, getting it hot, while asking nothing of the other seven (7). These applications are typically very expensive ones, so you might think that they would make use of them.
Oh, but no. Give me two cores, 100 cores, or anywhere in between. I, as a power-user, will actually never notice a difference.
Get the programmers to write MPA software. Only then will I think about believing the hype about multiple cores.
Liberal detected.
~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
I am waiting for the Mach30TurboLazer. Call me a Luddite. :)
~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
Price for performance?
The chips no matter how small their transistors are still need to fit on a standard die.
To get these chips to run faster you can add more transistors and/or better optimize them for their use.
When you focus on the latter there is lass of a case of running out of space.
If we buy a bigger home we don't buy bigger furniture just more of it.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Something fast enough to run Minecraft!
What IS really interesting is that the technology offered to the masses today is last years news, read: only what they want us to see.. I know for a fact that the big money has stuff we see in movies. Society is getting to the point where technology is literally climbing up our asses, which is why all the tremors over backdoors, encryption, and in general the mass confusion of information people are fighting over. Greed is quite literally become psychotic. News at 11
Non ironic Beowulf clusters.
Non impediti ratione cogitationus.
If Windows is now fast enough, Windows 11 will be out and this CPU will be minimum requirements.
So the product is Intel's not quite released compute accelerator, featuring new micro architecture, memory technology, and using the latest chip fab capabilities.
The most readily available competition with released numbers is an nVidia K80, a year old product using 5 year old memory technology, 5 year old chip fab capabilities, Set to be superseded by their refresh using state of the art fab, memory, and microarchitecture, which would actually compete toe to toe with what Intel announced.
This *should* make for an unambiguous trouncing of the nVidia product by Intel. So let's compare some metrics (not the best mechanism, but without real world numbers, settling for Rpeak and such).
Compute stands at 10TFlops SP and 2.9 TFlops DP on the K80, meaning Intel's brand new offering doesn't reach the SP performance of an 'ancient' product and barely edges them out on DP.
Memory capacity is actually lower than the K80 as well (16 GB intel v 24GB nvidia).
There is of course chance that even when going toe to toe with Pascal, that the ability to actually extract the promised performance will be better, but given how this doesn't unambiguously trump the K80 on paper, it's quite likely that Pascal will be overwhelming.
I also think the MIC product line will become redundant around Sky Lake dual socket time, when the main processor line starts having the AVX512 goodness.
XML is like violence. If it doesn't solve the problem, use more.
Sure! Here you go!
Society is getting to the point where technology is literally climbing up our asses,
So you are saying that "big money" has next generation butt plugs?
APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
http://starwars.wikia.com/wiki...
Are you sure you want to shave with one though, that might hurt...
APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
Yeah, my first thought on seeing that monster chunk of silicon: "Defectivity is going to make that thing expensive as hell."
Are they not using the trick of selling chips with a defective core as a lower core count (like the old Phenom X3). I assumed that was why you get strange numbers like 61 cores.
Maybe you could, you know, stop posting as an anonymous coward before you go on this rant. It might make it slightly easier to impress people. Probably not though, given the rather lame content.
It would certainly be a close shave. :)
~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
It would give a whole new meaning to razor burn.
APK likes to ask for responses to the same things over and over. Maybe he just likes the responses?
How about you go and fuck yourself you pathetic sack of shit
The new right fascists are bilingual. They speak English and Bullshit.
Another reason they may use the larger version is that they simply have much more room to work with. In a CPU there is a very small area that can be taken up by cores since so many things have been integrated these days like video processors and the north-bridge. Here they have an entire pci-e card to play with, so they can use the larger architecture without worrying about space.
This is merely my conjecture though.