that thing cannot even run tetris for two reasons, you won't have a GUI and secondly those parts appear only on presentations. I don't think that it will be an easy task to deliver this supercomputer, they only have the Infiniband ready :D
You do understand that it is SUPPOSED to be hard to deliver these machines, right?
The US government is trying to do more than one thing when it orders machines like this --- not just to get the machine itself, but ALSO to provide support for advancing US industry in one of the areas where it remains the world leader. That means stretch goals and an expectation that, occasionally, the deadline will slip or a plan B will be required if something turned out to be a little too ambitious.
"unified" is technical term. Its mean "no translation" (without coherence), or "same point of view over data" (with coherence).
"unified" as technical term thus do not mean "single thing"
"virtual" as technical term would mean that this "translation" is just hidden from developer. Right now Nvidia have such "virtual" unified memory for their CUDA. Its not unified but driver takes care of translation. Intel and some AMDs have unified memory without coherence. Memory can hold same values, and can be addressed by same methods, and can be seen by apps/os as single continuous address space. App developer just need to make sure that changed to data made in CPU are visible in GPU and vice versa.
AMD also have fully unified memory (hUMA), where data changed in CPU is instantly visible to GPU and vice versa without any app developer action.
"Speaking of NVLink, as IBM’s POWER family is the first CPU family to support NVLink it should come as no surprise that NVLink will be the CPU-GPU and GPU-GPU interconnect for these computers."
NVLink appears to be a variant of CAPI used in the POWER8 designs. They're both electrically PCIe with an encapsulated protocol on top.
What could change for the POWER9 is that IBM could move to optical links between processors which in turn would change CAPI. IBM, like Intel, has been researching silicon photonics for awhile now and this would be the first time they could produce a chip utilizing it. (IBM produced a massive optical hub for the HPC version of the POWER7 but it didn't use silicon photonics.)
The other amusing thing that this is the first big venture between IBM and nVidia since the Geforce 6800 (which IBM manufactured for nVidia). Times have certainly changed as both IBM and nVidia are now fabless companies.
NVLink is not electrically PCIe. It's a completely different signalling technology, more like what is described here: http://isscc.org/about/awards_2013.html
According to the press release, "over 90%" of peak flops come from the GPU. The PPC component is very obviously there for serial sections - Amdahl's Law can be a harsh mistress.
The reason why I want a PowerPC is that I buy it and 99% of you guys will not have it. And it's so powerful that I don't have to worry about roadmaps or updating it for a while. And, from the name "PowerPC," I should be able to run Crysis 3, and COD: Ghost at the same time. But, I know this will never happen. So, as long as a PowerPC is out there somewhere makes me hopeful somehow for some reason, which I don't know why.
Outside of console-specific games which could not run on it either, Crysis and CoD would have x86 binaries that would not run on this (unless you did binary translation, but even then, the latency would likely be too much for a game, even if you have all this power at your disposal it's largely single threaded).
So more likely, if you bought one, you'd be staring at a CLI display which looks like it's from the future the 70s conceived, and have absolutely no clue what to do with it by the sounds of things :P
Also you might already have "a powerpc", if you have an Xbox 360, Wii, Wii U, PS3, etc. They used PowerPC processors. Unless by "a powerPC", you meant the whole supercomputer shebang with thousands of powerPC processors in it, but doesn't make sense with the phrasing.
I do have a PowerPC ala 12" Apple Powerbook G4!!! Still works, too. And, stable as shit. The only time I arose to instabilities using my Mac and OS X was when I updated her with an intel Mac. Although, I am not an engineer or expert, so as to know what I am talking about. Just an observation from a pseudo-tech observer.
CLI Display? I guess I have always been fond of the word "Super Computers" since the 90's when there were RISC based workstations from Silicon Grafx running UNIX that I could never afford. Until, of course, I got a 12" G4 Apple Powerbook because it was cheaper than the 15" and had "RISC" cpu guts. Seemed cool then and seems even cooler now this Powerbook. So much stabler and seems not as sensitive as the new MBP's. I even dropped her a couple times and one big one when my shoulder bag broke while running to the train. It left a dent on her but she still ticks. I even disassembled her to try to bang the dent back into place from the inside, reapplied thermal paste and just out of curiosity and for its own sake since I got a 15" MBP unibody replacement. And I didn't even break her and still works to this day. The mid-2010 MBP on the other hand... is working now... but, oh boy, was she a headache. Hope she will last as long as her smaller sister.... but I'm rambling on now for too long. So, bye....
Yeah, usually on mainframes or headless supercomputers like this, what you'll see when you plug in a display or remote into one of the individual systems is something akin to this, very basic, not much put into the UI, but functional as heck and never a hiccup with all the power behind it.
Also those SGI "supercomputers" you were probably thinking of were desktop personal-ish computers, while the thing in the image above is the size of multiple rooms.
Something like this I'm assuming? I think the IBM Supercomputers would be larger than your house :P
Yep. They were desktops. Mid tower size. And, I think they were in grey, which were totally different than the beige window pc's or the off-white apple quadras....
i would assume that if I have one of these things, being an IBM Power9 running this architecture that I'd be able to interconnect as many pc boxes that I can afford, accordingly to the size of the room where the computers will be?....
At least. Actually I'm kind of surprised this will be coming online in 2017 since it's supposed to be based on Volta, which you wouldn't expect to see until 2018 given Nvidia's 2 year architecture schedule. In any case this is a long way out. We basically have 2 generations worth of Nvidia architectures to go before we hit Volta (big Maxwell ~2015, Pascal ~2016).
... actually this seems like a really aggressive (borderline unrealistic) schedule.
This is certainly a win for IBM/NVIDIA, but I wonder what it says about Cray/Intel? Has Cray lost its competitive edge? It would be one thing for LLNL to go for IBM, since they have traditionally used IBM supercomputers due to the work load they tend to run, but ORNL has most often procured Cray supercomputers. The problem is that Cray has been using x86 CPUs (AMD in the past, and now Intel), and Intel is certainly not keen on using NVLINK, or anything NV related, for that matter. That means Cray's only hope to stay competitive is for Xeon Phi to do better than NVIDIA Tesla. It doesn't look like they were able to make it in this round. Can they catch up and overtake IBM/NVIDIA? If not, the consequences for Cray could be very significant.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
29 Comments
Back to Article
xsoft7 - Monday, November 17, 2014 - link
Can it run Crisis?nathanddrews - Monday, November 17, 2014 - link
No, no, no... Can it run Assassin's Creed Unity?basroil - Monday, November 17, 2014 - link
It could probably optimize AC:U enough that it could run on your phone with the same quality... (though file sizes would be quite a bit larger)domboy - Monday, November 17, 2014 - link
I think a better question would be how many instances of <insert your preferred application/game> can it run at simultaneously??tipoo - Monday, November 17, 2014 - link
It can run crysis about as much as that joke is funny. Only in some far fetched theoretical.Yorgos - Monday, November 17, 2014 - link
that thing cannot even run tetris for two reasons,you won't have a GUI and secondly those parts appear only on presentations.
I don't think that it will be an easy task to deliver this supercomputer, they only have the Infiniband ready :D
name99 - Tuesday, November 18, 2014 - link
You do understand that it is SUPPOSED to be hard to deliver these machines, right?The US government is trying to do more than one thing when it orders machines like this --- not just to get the machine itself, but ALSO to provide support for advancing US industry in one of the areas where it remains the world leader. That means stretch goals and an expectation that, occasionally, the deadline will slip or a plan B will be required if something turned out to be a little too ambitious.
Samus - Monday, November 17, 2014 - link
300 PFLOPS, 10MW.Makes previous efficiency look like a 63 Lincoln Continental. The real problem though is how to handle the heat generated by 10MW of equipment.
Noëlius - Monday, November 17, 2014 - link
The same way you would cool a raised floor area about 18,000 sq/ft. That's a fraction of a large data centre! I don't see a problem.przemo_li - Wednesday, November 19, 2014 - link
By offering for no charge to heat whole neighborhood? Not only this helps with dissipating heat, but it also rise Performance/Watt still higher!jwcalla - Monday, November 17, 2014 - link
I think the "unified memory" will be virtual, of course.przemo_li - Wednesday, November 19, 2014 - link
"unified" is technical term. Its mean "no translation" (without coherence), or "same point of view over data" (with coherence)."unified" as technical term thus do not mean "single thing"
"virtual" as technical term would mean that this "translation" is just hidden from developer.
Right now Nvidia have such "virtual" unified memory for their CUDA. Its not unified but driver takes care of translation.
Intel and some AMDs have unified memory without coherence. Memory can hold same values, and can be addressed by same methods, and can be seen by apps/os as single continuous address space. App developer just need to make sure that changed to data made in CPU are visible in GPU and vice versa.
AMD also have fully unified memory (hUMA), where data changed in CPU is instantly visible to GPU and vice versa without any app developer action.
Kevin G - Monday, November 17, 2014 - link
"Speaking of NVLink, as IBM’s POWER family is the first CPU family to support NVLink it should come as no surprise that NVLink will be the CPU-GPU and GPU-GPU interconnect for these computers."NVLink appears to be a variant of CAPI used in the POWER8 designs. They're both electrically PCIe with an encapsulated protocol on top.
What could change for the POWER9 is that IBM could move to optical links between processors which in turn would change CAPI. IBM, like Intel, has been researching silicon photonics for awhile now and this would be the first time they could produce a chip utilizing it. (IBM produced a massive optical hub for the HPC version of the POWER7 but it didn't use silicon photonics.)
The other amusing thing that this is the first big venture between IBM and nVidia since the Geforce 6800 (which IBM manufactured for nVidia). Times have certainly changed as both IBM and nVidia are now fabless companies.
BryanC - Monday, November 17, 2014 - link
NVLink is not electrically PCIe. It's a completely different signalling technology, more like what is described here:http://isscc.org/about/awards_2013.html
tipoo - Monday, November 17, 2014 - link
Is it known how many of those gflops come from the CPU side or the GPU side on those?SarahKerrigan - Monday, November 17, 2014 - link
According to the press release, "over 90%" of peak flops come from the GPU. The PPC component is very obviously there for serial sections - Amdahl's Law can be a harsh mistress.tipoo - Monday, November 17, 2014 - link
Makes sense, I expected it to be hugely GPU-leaning, even though the Power8s are some absolutely bonkers processors (in a good way).wurizen - Monday, November 17, 2014 - link
The reason why I want a PowerPC is that I buy it and 99% of you guys will not have it. And it's so powerful that I don't have to worry about roadmaps or updating it for a while. And, from the name "PowerPC," I should be able to run Crysis 3, and COD: Ghost at the same time. But, I know this will never happen. So, as long as a PowerPC is out there somewhere makes me hopeful somehow for some reason, which I don't know why.tipoo - Monday, November 17, 2014 - link
Outside of console-specific games which could not run on it either, Crysis and CoD would have x86 binaries that would not run on this (unless you did binary translation, but even then, the latency would likely be too much for a game, even if you have all this power at your disposal it's largely single threaded).So more likely, if you bought one, you'd be staring at a CLI display which looks like it's from the future the 70s conceived, and have absolutely no clue what to do with it by the sounds of things :P
tipoo - Monday, November 17, 2014 - link
Also you might already have "a powerpc", if you have an Xbox 360, Wii, Wii U, PS3, etc. They used PowerPC processors. Unless by "a powerPC", you meant the whole supercomputer shebang with thousands of powerPC processors in it, but doesn't make sense with the phrasing.wurizen - Monday, November 17, 2014 - link
I do have a PowerPC ala 12" Apple Powerbook G4!!! Still works, too. And, stable as shit. The only time I arose to instabilities using my Mac and OS X was when I updated her with an intel Mac. Although, I am not an engineer or expert, so as to know what I am talking about. Just an observation from a pseudo-tech observer.wurizen - Monday, November 17, 2014 - link
CLI Display? I guess I have always been fond of the word "Super Computers" since the 90's when there were RISC based workstations from Silicon Grafx running UNIX that I could never afford. Until, of course, I got a 12" G4 Apple Powerbook because it was cheaper than the 15" and had "RISC" cpu guts. Seemed cool then and seems even cooler now this Powerbook. So much stabler and seems not as sensitive as the new MBP's. I even dropped her a couple times and one big one when my shoulder bag broke while running to the train. It left a dent on her but she still ticks. I even disassembled her to try to bang the dent back into place from the inside, reapplied thermal paste and just out of curiosity and for its own sake since I got a 15" MBP unibody replacement. And I didn't even break her and still works to this day. The mid-2010 MBP on the other hand... is working now... but, oh boy, was she a headache. Hope she will last as long as her smaller sister.... but I'm rambling on now for too long. So, bye....tipoo - Monday, November 17, 2014 - link
Yeah, usually on mainframes or headless supercomputers like this, what you'll see when you plug in a display or remote into one of the individual systems is something akin to this, very basic, not much put into the UI, but functional as heck and never a hiccup with all the power behind it.http://upload.wikimedia.org/wikipedia/commons/2/29...
tipoo - Monday, November 17, 2014 - link
Also those SGI "supercomputers" you were probably thinking of were desktop personal-ish computers, while the thing in the image above is the size of multiple rooms.Something like this I'm assuming? I think the IBM Supercomputers would be larger than your house :P
http://upload.wikimedia.org/wikipedia/commons/5/55...
wurizen - Tuesday, November 18, 2014 - link
Yep. They were desktops. Mid tower size. And, I think they were in grey, which were totally different than the beige window pc's or the off-white apple quadras....i would assume that if I have one of these things, being an IBM Power9 running this architecture that I'd be able to interconnect as many pc boxes that I can afford, accordingly to the size of the room where the computers will be?....
alxxx - Tuesday, November 25, 2014 - link
You can buy a ppc board from freescale or Applied microhttp://www.freescale.com/webapp/sps/site/homepage....
500MHz -1.8GHz
or buy a board with an older xilinx virtex2 fpga with 2 ppc hardcores
Embedded planet (and others element14) sell ppc sbc's
http://www.embeddedplanet.com/product/single-board...
Should no problems running linux(ubuntu, fedora ,deb) all have ports.
A very few set top boxes are ppc.
I think you mean you want a power not ppc or powerpc.
You could always run a ppc softcore in an fpga
mamisano - Monday, November 17, 2014 - link
All of this is still 2+ years away!dragonsqrrl - Tuesday, November 18, 2014 - link
At least. Actually I'm kind of surprised this will be coming online in 2017 since it's supposed to be based on Volta, which you wouldn't expect to see until 2018 given Nvidia's 2 year architecture schedule. In any case this is a long way out. We basically have 2 generations worth of Nvidia architectures to go before we hit Volta (big Maxwell ~2015, Pascal ~2016).... actually this seems like a really aggressive (borderline unrealistic) schedule.
Ktracho - Monday, November 17, 2014 - link
This is certainly a win for IBM/NVIDIA, but I wonder what it says about Cray/Intel? Has Cray lost its competitive edge? It would be one thing for LLNL to go for IBM, since they have traditionally used IBM supercomputers due to the work load they tend to run, but ORNL has most often procured Cray supercomputers. The problem is that Cray has been using x86 CPUs (AMD in the past, and now Intel), and Intel is certainly not keen on using NVLINK, or anything NV related, for that matter. That means Cray's only hope to stay competitive is for Xeon Phi to do better than NVIDIA Tesla. It doesn't look like they were able to make it in this round. Can they catch up and overtake IBM/NVIDIA? If not, the consequences for Cray could be very significant.