Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia to Acquire Arm for $40B

nvidianews.nvidia.com

Post reply on HN
Print view View original

321–330 of 784 posts

Re: Nvidia to Acquire Arm for $40B

#321

Post by StillBored »

dragontamer wrote: post #248
darksaints wrote: post #226

Earlier quoted context omitted.

Itanium was an excellent idea that needed investment in compilers. Nobody wanted to make that investment because speculative execution got them 80% of the way there without the investment in compilers. But as it turns out, speculative execution was a phenomenally bad idea, and patching its security vulnerabilities has set back processor performance to the point where VLIW seems like a good idea again. We should have…

NVidia Volta: https://arxiv.org/pdf/1804.06826.pdf Each machine instruction on NVidia Volta has the following information: * Reuse Flags * Wait Barrier Mask * Read/Write barrier index (6-bit bitmask) * Read Dependency barriers * Stall Cycles (4-bit) * Yield Flag (1-bit software hint: NVidia CU will select new warp, load-balancing the SMT resources of the compute unit) Itanium's idea of VLIW was commingled with other…

GPU's, aren't general purpose compute. EPIC did fairly well with HPC/etc style applications as well, it was everything else that was problematic. So, yes there are a fair number of workload and microarch decision similarities. But right now, those workloads tend to be better handled with a GPU style offload engine (or as it appears the industry is slowly moving, possibly a lot of fat vector units attached to a normal core).

Re: Nvidia to Acquire Arm for $40B

#322

Post by Ecco »

topspin wrote: post #154
walterbell wrote: post #33

Talking points from the founders of Arm & Nvidia: https://www.forbes.com/sites/patrickmoorhead/2020/09/13/its-... > Huang told me that first thing that the combined company will do is to, “bring NVIDIA technology through Arm’s vast network.” So I’d expect NVIDIA GPU and NPU IP to become available quickly to smartphone, tablet, TV and automobile SoC providers as quickly as possible. > Arm CEO Simon Segars framed it we…

My instincts are telling me this is smoke and mirrors to rationalize a $40E9 deal. The only part of that that computes at all is the GPU integration, and that only works if NVIDIA doesn't terrorize Arm licencees. The rest is buzzwords.

Want to be nerdy and use powers-of-ten? Fine by me! But then please go all the way: $4E10!!!

Re: Nvidia to Acquire Arm for $40B

#323

Post by shaklee3 »

sillysaurusx wrote: post #284
shaklee3 wrote: post #255

Earlier quoted context omitted.

I gave you an example of something you can't do, which is an overlap-save FFT, and you ignored that completely. Please implement it, or show me any example of someone implementing any custom FFT that's not a simple, standard, batched FFT. I'll take any example of implementing any type of signal processing pipeline on TPU, such as a 5G radio. Your last sentence is pretty funny: a GPU can't do certain workloads because…

Which part of this can't be done on TPUs? https://en.wikipedia.org/wiki/Overlap%E2%80%93save_method#Ps... As far as I can tell, all of those operations can be done on TPUs. In fact, I linked to the operation list that shows they can be. You'll need to link me to some specific implementation that you want me to port over, not just namedrop some random algorithm. Got a link to a github? If your point is "There isn't a…

Let's just leave this one alone then. I can't argue with someone who claims anything is possible, yet absolutely nobody seems to be doing what you're referring to (except you). A100 now tops all MLPerf benchmarks, and the unavailable TPUv4 may not even keep up.

Trust me, I would love if TPUs could do what you're saying, but they simply can't. There's no direct DMA from the NIC to where I can do a streaming application at 40+Gbps to it. Even if TPU could do all the things you claim, if it's not as fast as the A100, what's the point? To go through undocumented pain to prove something?

Re: Nvidia to Acquire Arm for $40B

#324

Post by dragontamer »

StillBored wrote: post #321
dragontamer wrote: post #248

Earlier quoted context omitted.

NVidia Volta: https://arxiv.org/pdf/1804.06826.pdf Each machine instruction on NVidia Volta has the following information: * Reuse Flags * Wait Barrier Mask * Read/Write barrier index (6-bit bitmask) * Read Dependency barriers * Stall Cycles (4-bit) * Yield Flag (1-bit software hint: NVidia CU will select new warp, load-balancing the SMT resources of the compute unit) Itanium's idea of VLIW was commingled with other…

GPU's, aren't general purpose compute. EPIC did fairly well with HPC/etc style applications as well, it was everything else that was problematic. So, yes there are a fair number of workload and microarch decision similarities. But right now, those workloads tend to be better handled with a GPU style offload engine (or as it appears the industry is slowly moving, possibly a lot of fat vector units attached to a normal…

I'm not talking about the SIMD portion of Volta.

I'm talking about Volta's ability to detect dependencies. Which is null: the core itself probably can't detect dependencies at all. Its entirely left up to the compiler (or at least... it seems to be the case).

AMD's GCN and RDNA architecture is still scanning for read/write hazards like any ol' pipelined architecture you learned in college. The NVidia Volta thing is new, and probably should be studied from a architectural point of view.

Yeah, its a GPU-feature on NVidia Volta. But its pretty obvious to me that this explicit dependency-barrier thing could be part of a future ISA, even one for traditional CPUs.

Re: Nvidia to Acquire Arm for $40B

#325

Post by spullara »

zdw wrote: post #306
AnthonyMouse wrote: post #175

Earlier quoted context omitted.

> The perpetual architecture license folks that make their own cores like Apple, Samsung, Qualcomm, and Fujitsu (I think they needed this for the A64FX, right?) will be fine There is one thing they would need to worry about though, which is that if the rest of the market moves to RISC-V or x64 or whatever else, it's not implausible that someone might at some point make a processor which is superior to the ones those…

> Their best move might be to forget about the architecture license and make the switch to something else with the rest of the market. This assumes that there isn't some other factor in transitioning architecture - this argument could boil down in the mid 2000's to "Why not go x86/amd64", but you couldn't buy a license to that easily (would need to be 3-way with Intel/AMD to further complicate things) Apple has done…

They did dump ZFS when they decided they didn't like the licensing terms.

Re: Nvidia to Acquire Arm for $40B

#326

Post by na85 »

Ecco wrote: post #320
na85 wrote: post #274

Earlier quoted context omitted.

I miss DIP chips that would fit on breadboards. I don't have steady enough hands to solder QFP onto a PCB, and I'm too cheap to buy an oven :(

You’re supposed to drag-solder those. Look it up on YouTube, it’s super easy. The hardest part is positioning the chip, but it’s actually easier than with an oven, because you can rework it if you only solder one or two pins :)

Wow, just looked up a video and some guy did an 0.5mm pitch chip pretty darn quickly. Thank you!

Re: Nvidia to Acquire Arm for $40B

#327

Post by dragontamer »

anarazel wrote: post #319
dragontamer wrote: post #303

Earlier quoted context omitted.

VLIW was the best implementation (20 years ago) of instruction level parallelism. But what have we learned in these past 20 years? * Computers will continue to become more parallel -- AMD Zen2 has 10 execution pipelines, supporting 4-way decode and 6-uop / clock tick dispatch per core, with somewhere close to 200 registers for renaming / reordering instructions. Future processors will be bigger and more parallel, Ice…

I don't understand how an increase, including the implied variability, of CPU internal parallelism and VLIW benefits go together?

I'm talking about a mythical / mystical VLIW architecture. Obviously, older VLIW designs have failed in this regards... but I don't necessarily see "future" VLIW processors making the same mistake.

Perhaps from your perspective, a VLIW architecture that fixes these problems wouldn't necessarily be VLIW anymore. Which... could be true.

Re: Nvidia to Acquire Arm for $40B

#328

Post by mkl »

scarface74 wrote: post #223
ranbumo wrote: post #65

Earlier quoted context omitted.

Yes. It'd have been reasonable to block sales to non eu parties for national security reasons. Now arm is yet another US company.

Isn’t the UK leaving the EU?

Yes, but the Brexit referendum was only 1 month before SoftBank acquired Arm Holdings. The deal was probably already in progress, and finalised before the UK had any real policies about Brexit, so EU requirements would have been reasonable if decided ahed of time. But the timing may also explain the lack of any national security focused requirement (general confusion).

Re: Nvidia to Acquire Arm for $40B

#329

Post by walterbell »

btown wrote: post #315
xbmcuser wrote: post #282

Earlier quoted context omitted.

Are they getting 10% of nvidia or keeping 10% of Arm

From the press release directly, it appears to be 10% of NVIDIA: https://nvidianews.nvidia.com/news/nvidia-to-acquire-arm-for... > Under the terms of the transaction, which has been approved by the boards of directors of NVIDIA, SBG and Arm, NVIDIA will pay to SoftBank a total of $21.5 billion in NVIDIA common stock and $12 billion in cash, which includes $2 billion payable at signing. The number of NVIDIA shares to…

> no new entity is mentioned in the press release itself

The Forbes article is based on a joint interview today with the CEOs of Arm and Nvidia, who could have provided more detail than the press release, specifically:

> Arm operating structure: Arm will operate as an NVIDIA division

This level of detail can be confirmed during the analyst call on Monday. Operating as a separate division would help assuage concerns about Arm's independence. The press release says:

> Arm will remain headquartered in Cambridge ... Arm’s intellectual property will remain registered in the U.K.

Those statements are both consistent with Arm operating as a UK-domiciled business that is owned by Nvidia.

Re: Nvidia to Acquire Arm for $40B

#330

Post by sillysaurusx »

shaklee3 wrote: post #323
sillysaurusx wrote: post #284

Earlier quoted context omitted.

Which part of this can't be done on TPUs? https://en.wikipedia.org/wiki/Overlap%E2%80%93save_method#Ps... As far as I can tell, all of those operations can be done on TPUs. In fact, I linked to the operation list that shows they can be. You'll need to link me to some specific implementation that you want me to port over, not just namedrop some random algorithm. Got a link to a github? If your point is "There isn't a…

Let's just leave this one alone then. I can't argue with someone who claims anything is possible, yet absolutely nobody seems to be doing what you're referring to (except you). A100 now tops all MLPerf benchmarks, and the unavailable TPUv4 may not even keep up. Trust me, I would love if TPUs could do what you're saying, but they simply can't. There's no direct DMA from the NIC to where I can do a streaming applicatio…

FWIW, you can stream at 10Gbps to TPUs. (I've done it.)

10Gbps isn't quite 40Gbps, but I think you can get there by streaming to a few different TPUs on different VPC networks. Or to the same TPU from different VMs, possibly.

The point is that there's a realistic alternative to nVidia's monopoly.

Post reply on HN