[HN Gopher] The first two custom silicon chips designed by Micro...
       ___________________________________________________________________
        
       The first two custom silicon chips designed by Microsoft for its
       cloud
        
       Author : buildbot
       Score  : 104 points
       Date   : 2023-11-15 16:02 UTC (6 hours ago)
        
 (HTM) web link (www.theverge.com)
 (TXT) w3m dump (www.theverge.com)
        
       | spandextwins wrote:
       | So does NVIDIA, just turns out NVIDIA can profit more because of
       | their software and the software ecosystem around it adds so much
       | value, nobody can compete. It's gonna take a lot of work and many
       | years to approach that. Even by leveraging AI. And by diverging
       | with their own chips, they're gonna miss out on the mainstream.
        
         | ugh123 wrote:
         | I don't know much about the software-side of nvidia/gpus +
         | LLMs. Can you catch me up on what software they've created
         | means as a differentiator? Is that CUDA? How does this relate
         | to things like tensorflow with Google's chips?
        
       | jedberg wrote:
       | It was only a matter of time. Google announced theirs years ago,
       | Amazon announced theirs last year.
       | 
       | Right now NVIDIA has the lead because they have the better
       | software, but they can't make the chips fast enough. Will be
       | interesting to see if their better software continues to keep
       | them in the lead or if people are more interested in getting the
       | capacity in any form.
        
         | rehitman wrote:
         | I would say software is very important. We already have tons of
         | different models, standards, library, etc. I usually have
         | smooth experience if I am using NVDA, but any other variation,
         | I have to spend some times to get things started.
         | 
         | Supporting the common librarires that I use is very important
         | for me to chose the cloud platform.
        
         | Jagerbizzle wrote:
         | Considering that Jensen is on stage with Satya at the moment
         | sharing the keynote of Microsoft Ignite, I suspect NVIDIA won't
         | be going anywhere anytime soon.
        
           | paulpan wrote:
           | A bit surprised to see Jensen's stage appearance since
           | clearly Microsoft's success with its own AI chips means less
           | business for Nvidia's chips.
           | 
           | Because different than the ARM chip also announced in the
           | same Ignite event, Microsoft doesn't exactly "need" nor can
           | fully utilize an AI chip. Google trains its foundational
           | models (e.g. Gemini) on its own TPU hardware but Microsoft's
           | is heavily reliant on OpenAI for its generative AI serving
           | needs.
           | 
           | Unless Microsoft is planning to acquire OpenAI fully and
           | switch over from Nvidia hardware...
        
             | echelon wrote:
             | > Unless Microsoft is planning to acquire OpenAI fully
             | 
             | They're going to play a modified version of the old
             | Rareware trick.
             | 
             | It's also a pretty great game to buy up OpenAI equity,
             | which ultimately gets spent on Microsoft compute. Two
             | birds, one stone.
        
               | WanderPanda wrote:
               | I don't think the margins here are so considerable that
               | we can assume revenue = profit
        
               | echelon wrote:
               | They can be losing money, but gaining in market share and
               | moat.
               | 
               | Almost nobody in this game cares about profit right now.
        
               | kmeisthax wrote:
               | ...they're going to overbid on a studio that was actively
               | falling apart, after being rebuffed from buying one of
               | the biggest giants in the business[0], all as part of an
               | ill-advised attempt to muscle into a game business they
               | didn't understand?
               | 
               | [0] Microsoft tried to buy Nintendo very early on
        
             | aseipp wrote:
             | Microsoft absolutely runs their own models on their own
             | hardware, at scale, and they have done so for years just
             | like every other hyperscaler -- Project Brainwave was first
             | publicly talked about as far back as 2018. The generative
             | LLM craze is a recent phenomenon in comparison. They are
             | absolutely going to go all in on putting AI functionality
             | in Bing, in Excel, in Windows, etc etc. To do that, you
             | need hardware.
             | 
             | None of this is really strange. It also wasn't strange when
             | Google announced H100 systems while also pushing TPUs they
             | developed. Microsoft has Jensen on stage because customers
             | of Microsoft Azure demand Nvidia products. Customers of
             | Google Cloud demand Nvidia products. So, they provide them
             | those products, because not providing them loses those
             | customers. It's that simple. Everyone involved in these
             | deals acknowledges this.
        
               | buildbot wrote:
               | Awesome! Someone who knows about Brainwave!
        
               | aseipp wrote:
               | Yeah! Did/do you work on it? The original publications
               | were good timing; I was working as a consultant on an
               | FPGA-based ML accelerator at the time the original stuff
               | was talked about, and I really enjoyed reading everything
               | I could about Brainwave! Really neat project from both a
               | system design perspective (e.g. heterogeneous compiler)
               | to the choice of using and interconnecting FPGAs and
               | integrating the network/software/ML stack (IIRC, there
               | was a good paper on the overlay network they used to make
               | those custom functions available on the global network
               | fabric.)
               | 
               | I'm guessing at this point the ASICs make a lot more
               | economic sense, though. :)
        
         | rapsey wrote:
         | nVidia still has the monopoly on training. Everyone else is
         | just making chips for inference.
        
           | muro wrote:
           | https://cloud.google.com/tpu/docs/training-on-tpu-pods
        
           | adrian_b wrote:
           | While NVIDIA provides the best absolute performance for
           | training, Intel (i.e. Gaudi) already provides much better
           | training performance per dollar.
           | 
           | The funny thing is that this fact has been shown
           | inadvertently by NVIDIA:
           | 
           | https://www.servethehome.com/nvidia-shows-intel-
           | gaudi2-is-4x...
        
             | sgillen wrote:
             | That article shows that it takes about 50x as long to train
             | gpt-3 with intel's offering vs Nvidia. At least in the
             | current environment, if you are training llms I think
             | almost no amount of cost savings can justify that.
        
               | adrian_b wrote:
               | That 50X is only if you can afford one thousand NVIDIA
               | H100.
               | 
               | There cannot be more than a handful of companies in the
               | entire world that could afford such a huge price (tens of
               | millions of $).
               | 
               | In comparison with a still extremely expensive cluster of
               | 64 NVIDIA H100, the difference in speed would reduce to
               | only two to three times, and paying several times less
               | for the entire training becomes very attractive.
        
           | jedberg wrote:
           | Amazon announced Trainum last year and recently announced
           | their partnership with Anthropic, where Claude will be
           | trained on Trainium.
           | 
           | https://press.aboutamazon.com/2023/9/amazon-and-anthropic-
           | an...
        
         | belval wrote:
         | > Amazon announced theirs last year
         | 
         | Inferentia (inf1) was GA'ed in December 2019 so it's actually
         | almost 4 years old now. The trainium (trn1) chips and the
         | Inferentia 2 (inf2) refresh is indeed 1 year old though.
        
           | jedberg wrote:
           | I was referring to trainium.
        
           | ShamelessC wrote:
           | Yikes those names are horrible.
        
             | bonton89 wrote:
             | They'll sound better after we hear Microsoft's names.
        
               | cameronh90 wrote:
               | Microsoft Azure Inference for Cloud Apps 365 Pro Live
               | Series X
        
               | Someone wrote:
               | I think you mean
               | 
               | Microsoft Azure(tm) Inference for Cloud Apps(c) 365
               | Pro(r) Live Series X(tm)(r)
        
               | mdaniel wrote:
               | I'm still _so bitter_ about this nonsense
               | https://www.microsoft.com/en-
               | us/security/business/identity-a...
               | 
               | > Azure Active Directory is now Microsoft Entra ID
               | 
               | ok, geez, thanks
        
               | bee_rider wrote:
               | I don't really know what an active directory is, but I
               | assume that the default type of directory is a passive
               | one, in that it just holds files or subdirectory (it
               | doesn't act). An active directory sounds like a directory
               | that is going to play tricks on me.
               | 
               | Entra ID sounds like a type of ID.
               | 
               | I'm not sure how something could legitimately have each
               | of these names. I assume the functionality changed pretty
               | dramatically over the lifespan of the product?
        
               | fragmede wrote:
               | Active Directory is Microsoft's database targeted mostly
               | for user login and associated data.
        
             | belval wrote:
             | Why? Inferentia => inference, trainium => training. Given
             | the usually naming of AWS product, having one where the
             | name roughly matches what it does is pretty good?
             | 
             | TPU is pretty good but is associated with Google. MTIA is
             | an acronym but still maps to what the chip does. ~~"Cobalt"
             | is worse as it does not mean anything~~ . Cobalt is the CPU
             | chip, MAIA is the accelerator so this matches Meta's
             | naming.
        
               | ShamelessC wrote:
               | > Why? Inferentia => inference, trainium => training.
               | 
               | Funny, that's precisely why I think the names are bad.
               | It's like if Google had chosen "Search-ola" as their
               | name. Way too on the nose and/or lazy. Having said that,
               | I don't really care all that much and I imagine that may
               | have been the spirit of those who chose the names.
        
               | mdaniel wrote:
               | heh, as someone who has to deal with this nonsense all
               | day <https://aws.amazon.com/products/> I would for sure
               | welcome some straightforward naming. $(echo "AWS Fargate"
               | | sed s/Fargate/ServerlessContainerium/)
        
               | organsnyder wrote:
               | My previous role was a lot of AWS, and I became convinced
               | that the value of an AWS cert was mostly learning how to
               | map all of the product names to their actual functions.
        
             | esafak wrote:
             | They're searchable and self-explanatory so they're not bad.
        
           | coredog64 wrote:
           | Graviton CPU is a year older.
        
         | scarface_74 wrote:
         | If they "can't make the chips fast enough" being TSMC's second
         | highest volume customer behind Apple and probably second in
         | priority, what chance does Microsoft have getting enough of
         | TSMCs capacity?
        
           | donatzsky wrote:
           | Microsoft does have the benefit of not having any customers
           | other than themselves, so volumes are smaller.
        
           | tambre wrote:
           | They're more constrained by advanced packaging (CoWoS)
           | capacity rather than the manufacturing of the silicon.
        
             | tyfighter wrote:
             | But from the pictures, the Maia 100 is also using CoWoS
             | packaging. That will be necessary for any new HBM chip.
        
       | kristianp wrote:
       | https://archive.is/jzxDG
        
       | pseudosavant wrote:
       | Not a lot of information about the chips yet. About 100B
       | transistors in the AI chip. For comparison, an RTX 4090 has 76B,
       | and an H100 has about 80B. So the Maia chip is pretty massive.
        
         | hulitu wrote:
         | This is like performance review based on written line of code.
        
           | Slartie wrote:
           | GPUs (and AI chips) are highly parallel, containing thousands
           | upon thousands of the same compute units. The performance of
           | these chips is very much dependent on having a sheer number
           | of transistors to form into as many compute units as
           | possible.
           | 
           | If we assume that Microsoft is roughly able to architect
           | compute units of a similar performance-to-number-of-
           | transistors ratio as nVidia is, then having twice the number
           | of transistors should roughly result in twice the
           | performance.
           | 
           | That is very different than it is with typical software. If
           | you give a programmer who needs to write 100 lines of code to
           | solve a given problem 100 more lines to fill, he won't simply
           | be able to copy-paste his 100 lines another time and by that
           | action be twice as fast at solving whatever problem you
           | tasked him with. With GPU compute units, such copy-pasting of
           | compute units is exactly what's being done (at least until
           | you hit the limits of other resources such as management
           | units, memory bandwidth etc.).
        
       | virgildotcodes wrote:
       | This seems a little unclear to me. They're saying this isn't
       | meant to compete with Nvidia, and that it's more of an ARM based
       | CPU?
       | 
       | So more of an SOC a la AWS graviton or Apple Silicon than a pure
       | GPU?
        
         | jsnell wrote:
         | There's two separate chip announcements here. One is a ML
         | accelerator (Maia). The other is an ARM CPU (Cobalt).
        
       | conradev wrote:
       | "Microsoft gave few technical details that would allow gauging
       | the chips' competitiveness versus those of traditional
       | chipmakers"
       | 
       | Clearly. All I got was "using ARM IP" and "TSMC N5"
        
         | buildbot wrote:
         | Also the new MX datatypes
         | (https://www.opencompute.org/blog/amd-arm-intel-meta-
         | microsof...), from the article:
         | 
         | "Manufactured on a 5-nanometer TSMC process, Maia has 105
         | billion transistors -- around 30 percent fewer than the 153
         | billion found on AMD's own Nvidia competitor, the MI300X AI
         | GPU. "Maia supports our first implementation of the sub 8-bit
         | data types, MX data types, in order to co-design hardware and
         | software," says Borkar. "This helps us support faster model
         | training and inference times.""
        
           | conradev wrote:
           | Good catch! That is very cool
        
       | minedwiz wrote:
       | Looking over another story on this at
       | https://www.zdnet.com/article/microsoft-unveils-first-ai-chi...,
       | Cobalt seems to be a general purpose ARM CPU.
        
       | monlockandkey wrote:
       | Arm in inevitable for the server. It's interesting how now days,
       | efficiency/power consumption is a consideration over pure raw
       | performance.
        
         | wargames wrote:
         | I genuinely don't see how x86 architecture will continue to
         | survive the next 10 years. It will of course take longer to
         | change home desktop users to new architectures; they will be
         | the last segment to switch, but it seems all but inevitable.
         | 
         | BTW, I'm not even speaking to whether x86 can compete at the
         | same power per watt... I think it just won't make sense
         | financially to be out of sync with the industry.
        
           | jimmaswell wrote:
           | I care vastly more about raw performance than energy usage
           | for my home systems. I also have good reasons to care about
           | the best single core performance. I don't see x86 going away
           | that fast.
        
             | magnio wrote:
             | FWIW, the current king of single core Geekbench is the M3
             | chips. Even the base M3 scores as high as the i9-14900K and
             | higher than the Ryzen 9 7950X3D, at less than half their
             | TDPs.
        
             | monlockandkey wrote:
             | Mobile, desktop, laptop, edge, server. These are the
             | domains of compute. 4 out of the 5 domains value power
             | efficiency. Laptop that were once x86 are now coming round
             | to Arm because it really does make a better product i.e
             | battery life and thermals. For the server, savings in
             | energy and cost of chip manufacturing, datacentres and
             | users both benefit.
        
           | adrian_b wrote:
           | Even if it should be possible to design Arm CPUs competitive
           | with the x86 CPUs, there are a lot of application domains for
           | which no vendor of Arm CPUs has ever attempted to make
           | competitive Arm CPUs.
           | 
           | For example, for scientific computation and computer-aided
           | design, Fujitsu is the only company that has designed Arm
           | CPUs that can compete with the x86 CPUs, but they do not sell
           | their CPUs on the free market.
           | 
           | For a huge company, the floating-point performance of the
           | CPUs is less important, because they can use datacenter GPUs
           | with even greater throughput, so the existing Arm server CPUs
           | could be good enough even for a supercomputer, as they only
           | have to move the data to and from the GPUs. However the small
           | businesses and the individuals cannot use datacenter GPUs,
           | which have huge prices, so they can use only x86 CPUs and
           | there is not the slightest chance of any alternative that
           | would appear soon.
           | 
           | Another application domain for which no Arm vendor has ever
           | made competitive devices is for cheap personal computers.
           | 
           | Nothing what Apple does matters, because they do not sell
           | computers, they only lend computers that remain under their
           | control and which are much more expensive than their
           | alternatives anyway.
           | 
           | Besides Apple, only Qualcomm, Mediatek and NVIDIA are able to
           | make Arm CPUs with a performance similar to the cheapest of
           | the Intel and AMD CPUs, but all these 3 companies demand for
           | their CPUs prices that are several times higher than the
           | prices of comparable x86 CPUs.
           | 
           | Like for CPUs with high floating-point or big integer
           | performance, there is not the slightest chance for the
           | appearance of any company that would be willing to sell Arm
           | CPUs that are both cheap and fast.
           | 
           | Also for server CPUs, all the companies that have attempted
           | to design Arm-based server CPUs have never designed models
           | suitable for small businesses or individuals, but only models
           | that can be bought only by very big companies.
           | 
           | I would not mind to switch from x86 to Arm, but there is
           | absolutely no perspective for that.
           | 
           | If the x86 CPUs would disappear, that would be a catastrophe
           | for the people who do not want to depend on the mercy of the
           | big companies. That would be a return to the times from
           | before the personal computers, when all computing had to be
           | done remotely, in the computing centers of big companies,
           | which have been renamed now as "clouds".
        
             | wmf wrote:
             | Grace is an ARM HPC CPU.
             | 
             | I agree that Qualcomm/Mediatek/Rockchip/Nvidia pricing is
             | really terrible but I guess prices don't matter when
             | there's almost no demand anyway.
        
           | hulitu wrote:
           | > I genuinely don't see how x86 architecture will continue to
           | survive the next 10 years.
           | 
           | ARM is ok only for reasonable performance at low power (if we
           | forget about VIA).
        
         | bee_rider wrote:
         | Most of the nodes I see every day are still x86. But I'm in an
         | academic environment, maybe things are slower over here. Does
         | ARM actually seem to have legs outside? (Other than, like,
         | nodes subsidized by Amazon's wish to in-house everything they
         | can).
        
           | monlockandkey wrote:
           | It's going to take time, but momentum is seriously starting
           | to build up now. Laptop market going to pick up with
           | Snapdragon X and cloud providers are going to continue with
           | more powerful designs.
        
             | bacchusracine wrote:
             | But will these run Linux, run AI stuff the way the Apple
             | Silicon seems to be able to do?
             | 
             | Because right now I'm looking to save up for a majorly
             | spec'd Apple MacbookPro just to be able to do this stuff on
             | a *nix operating system. I have no great love for Apple but
             | the abilities of their chips and the vast software
             | offerings are tempting this Linux guy in that direction.
             | 
             | Something that Microsoft cannot seem to do any more. I used
             | Windows from 3.x-WinME; NT3.51-WinXP, getting off before
             | Vista. What I've seen since then has done nothing to tempt
             | me back to their side. Since I unfortunately must deal with
             | Windows 10 at work, it definitely reinforces my distaste
             | for their systems....
             | 
             | So despite thinking OSX has been rendered ugly for the past
             | ten years now, I'm still thinking heavily in that
             | direction, even with the high costs. Snapdragon X sounds
             | nice enough but I have zero expectations based on past
             | behavior at those getting decent Linux support any time
             | soon. And no one else seems to even be trying, that one
             | Thinkpad aside.
        
               | bee_rider wrote:
               | Microsoft has taken a couple swings at making ARM laptops
               | (which, we should note, doesn't appear to be what they
               | are announcing here).
               | 
               | I'd expect a future hypothetical Microsoft ARM laptop to
               | be like a surface-RT; some Windows dropped on a third
               | party ARM chip. Microsoft is a software company, after
               | all. So it is more of a matter of, do they happen to have
               | bought a chip that supports Linux (probably yes, because
               | what hardware manufacturer wants to be dependent on one
               | company for OS support?) and can you get past Secureboot
               | (probably yes, after a couple years at least, when the
               | jailbreak happens).
        
         | swozey wrote:
         | I converted my corp apps to ARM (Fargate Graviton) last year
         | and our AWS bill plummeted and the time to fully initialize a
         | container did so as well.
         | 
         | I'd never tell the higher ups this but it was pretty easy, too.
         | I'll let them bask in my glory of saving the company
         | $60k/month.
        
       | andy_xor_andrew wrote:
       | Damn, Microsoft is playing all sides of this, huh.
       | 
       | Within a ten minute window:
       | 
       | - Satya announced GPT-4 runs (at least partly) on a new AMD
       | offering
       | 
       | - Satya announced an in-house chip for ML acceleration
       | 
       | - Satya brings NVidia CEO Jensen Huang on stage
       | 
       | they've got every horse in the race, huh
       | 
       | (disclaimer, I work for MS but all the stuff talked about here
       | far is waaaay above my paygrade haha, and all brand new info to
       | me)
        
         | onlyrealcuzzo wrote:
         | Microsoft really likes being valued as a growth stock...
         | 
         | Obviously they're going to play every angle.
        
         | kyboren wrote:
         | > - Satya brings NVidia CEO Jensen Huang on stage
         | 
         | This should be ringing alarm bells at FTC and DoJ.
         | 
         | You know what's even better than trust busting and breaking up
         | cartels? Preventing the formation of cartels and trusts in the
         | first place.
        
           | scarface_74 wrote:
           | Raising alarm bells because Microsoft is actually dealing
           | with multiple companies?
           | 
           | Would it be better for competition if Microsoft only used one
           | supplier?
        
       | aiman3 wrote:
       | NVIDIA should start offer AI cloud, buy DigitalOcean,
       | counterattack microsoft, google, amazon, since they are attacking
       | NVIDIA's territory.
        
         | stanac wrote:
         | They already have gaming cloud, wouldn't be unthinkable to
         | offer gpgpu cloud.
        
           | solardev wrote:
           | It's a hell of a cloud too. Geforce Now performs several
           | times better than Microsoft's crappy offering, xCloud (or
           | whatever it's called now).
           | 
           | Nvidia made some really amazing strides in the past few
           | years, taking over cloud gaming where Onlive and Stadia
           | utterly failed, making DLSS, etc.
           | 
           | I just hope they don't abandon us gamers for their AI stuff
           | :( Probably the entire gaming market is way smaller than the
           | potential AI market, just hopefully not too small to matter.
        
       | gary_0 wrote:
       | > Microsoft said it does not plan to sell the chips
       | 
       | Add it to the list of things you can't buy at any price, and can
       | only rent. That list is getting pretty long, especially if you
       | count "any electronic device you can't fully control or modify".
        
         | ronsor wrote:
         | Google does the same thing with their TPUs. The masses will be
         | left with the NVidia monopoly, while large companies will be
         | able to free themselves from that.
        
           | wmf wrote:
           | MI300 is coming.
        
             | machinekob wrote:
             | This time AMD for sure will fight with NV (its only failed
             | 20 times already copium)
        
               | wmf wrote:
               | On one hand this is a fair prediction but Triton exists
               | now and it didn't exist last time.
        
             | FuriouslyAdrift wrote:
             | December 6 launch date:
             | 
             | https://ir.amd.com/news-events/press-
             | releases/detail/1168/am...
        
           | llm_nerd wrote:
           | nvidia is a $1.2 trillion dollar company (the 6th largest
           | company by cap), and at this point AI is a huge component of
           | that wealth. It has appreciated by 3.3x since just the
           | beginning of this year.
           | 
           | If any of these companies _truly_ made competitive silicon
           | they absolutely would commercialize it.
           | 
           | I suspect they aren't as competitive as the press releases
           | hold them to be, and this Microsoft entrant is likely to
           | follow the same path. Like Google, Tesla, Amazon and others
           | it seems mostly an initiative to negotiate discounts from
           | nvidia.
           | 
           | It would be great if there were really competition. When
           | Google was hyped about their Tensor chips they did have a
           | period where they were looking to commercialize it, and there
           | are some pretty crappy USB products they sell.
        
             | jsnell wrote:
             | They are commercializing the silicon, by selling access to
             | it on their clouds.
             | 
             | Now, I know that what you actually mean is selling the
             | chips themselves to third parties :) But it's not obvious
             | that there's any point to it given their already existing
             | model of commercializing the chips.
             | 
             | First, literally everyone is already supply-constrained due
             | to limits on high end foundry capacity. Nvidia has a ton of
             | capacity because they're one of TSMC's top two customers.
             | The big tech companies will have much smaller allocations
             | which are used up just supplying their own clouds. Even if
             | the demand for buying these chips rather than renting were
             | there, they just don't have the chips to sell without
             | losing out on the customers who want to rent capacity.
             | 
             | Second, the chips by themselves are probably not all that
             | useful. A lot of the benefit is coming from the
             | silicon/system/software co-design. (E.g. the TPUv4 papers
             | spent as much attention on the optical interconnect as the
             | chips). Selling just chips or accelerator cards wouldn't do
             | much good to any customers. Nor can they just trust that
             | systems integrators could buy the cards and build good
             | systems to house them in. They need to sell and support
             | massive large scale custom systems to third parties. That's
             | not a core competency for any of them, it'll take years to
             | build up that org if you start now. And it means they need
             | to ship the software to the customers, it can't continue
             | being the secret sauce any more.
             | 
             | Nvidia on the other hand has been building up an ecosystem
             | and organization for exactly this for the last decade.
        
               | scarface_74 wrote:
               | > Nvidia has a ton of capacity because they're one of
               | TSMC's top two customers. The big tech companies will
               | have much smaller allocations which are used up just
               | supplying their own clouds.
               | 
               | And TSMCs top customer is not even playing in the cloud
               | space.
        
           | aleph_minus_one wrote:
           | > The masses will be left with the NVidia monopoly, while
           | large companies will be able to free themselves from that.
           | 
           | My bet: if it really becomes clear what capabilities an AI
           | accelerator chip needs _and_ lots of people want to run (or
           | even train) AIs _on their own_ computers, AI accelerators
           | will appear at the market. This is how capitalism typically
           | works.
           | 
           | My further bet: these AI accelerators will initially come
           | from China.
           | 
           | Just look at the history of Bitcoin: initially the blocks
           | were mined on CPUs, but then the miners switched to GPUs and
           | "everybody" was complaining about increasing GPU prices
           | because of all the Bitcoin mining. At some moment, Bitcoin
           | mining ASICs appeared from China and after those spread, GPUs
           | were not attractive anymore for Bitcoin mining (of course the
           | cryptocurrency fans who bought the GPUs for mining attempted
           | to use their investment for mining other cryptocurrencies).
        
             | brucethemoose2 wrote:
             | The capital costs are enormous, not even counting the CUDA
             | moat. It takes years to start producing a big AI processor.
             | 
             | Yet many startups and existing designers anticipated this
             | demand correctly, years in advance, and they are all
             | _still_ kinda struggling. Nvidia is massively supply
             | constrained. AI customers would be buying up MI250s, CS-2s,
             | IPUs, Tenstorrent accelerators, Gaudi 2s and so on en masse
             | if they wanted to... But they are not, and its not going to
             | get any easier once the supply catches up.
             | 
             | Unless there's a big one in stealth mode, I think we are
             | stuck with the hardware companies we have.
        
               | solardev wrote:
               | Is there not a distributed computing potential here like
               | there was for crypto mining? Some sort of seti@home/boinc
               | like setup where home users can donate or sell compute
               | time?
        
               | latchkey wrote:
               | The capex/opex is different for ML/AI than it was for
               | crypto mining. Totally different hardware profiles.
        
               | fragmede wrote:
               | you can setup a computer and sell time on it on a couple
               | of saas platforms, but only for inference. for training,
               | the slowness of the interconnect between nodes become a
               | bottleneck.
        
               | solardev wrote:
               | I see, thanks for the explanation!
        
               | brucethemoose2 wrote:
               | Yes, see projects like the AI Horde and Petals. I highly
               | recommend the Horde in particular.
               | 
               | Theres also some kind of actual AI crypto project that I
               | wouldn't touch with a 10 foot pole.
               | 
               | But ultimately, even if true distribution like Petals
               | figures out the inefficiency (and thats hard), it had the
               | same issue as non Nvidia hardware: its not turnkey.
        
               | aleph_minus_one wrote:
               | > Yet many startups and existing designers anticipated
               | this demand correctly, years in advance, and they are all
               | _still_ kinda struggling.
               | 
               | As I already hinted in my post: I see a huge problem in
               | the fact that in my opinion it still is not completely
               | clear to this day which capabilities an AI accelerator
               | really needs - too much is in my opinion still in a state
               | of flux.
        
               | brucethemoose2 wrote:
               | The answer is kinda "whatever Nvidia implements."
               | Research papers literally build around their hardware
               | capabilities.
               | 
               | A good example of this is Intel canceling, and AMD
               | sidelining, their unified memory CPU/GPU chips for AI.
               | They are super useful!.. In theory. But actually, they
               | totally useless because no one is programming frameworks
               | with unified memory SoCs in mind, as Nvidia does not make
               | something like that.
        
             | tomcam wrote:
             | > My bet: if it really becomes clear what capabilities an
             | AI accelerator chip needs and lots of people want to run
             | (or even train) AIs on their own computers, AI accelerators
             | will appear at the market.
             | 
             | My bet: in 6 months jart will have models running on local
             | or server, with support for all platforms and using only
             | 88K of ram ;)
        
         | layer8 wrote:
         | Sounds like an opportunity for hardware startups.
        
           | wmf wrote:
           | There are over a dozen AI hardware startups.
        
             | layer8 wrote:
             | Sounds like they are taking the opportunity.
        
             | Tempest1981 wrote:
             | Waiting to be assimilated by the big guys
        
           | scarface_74 wrote:
           | Hardware startups aren't going to stand a chance because they
           | have to fight for the scraps of capacity that is left over
           | after Apple and NVidia and the cloud providers use what they
           | can.
        
         | cjdrake wrote:
         | You can buy the Ampere chips:
         | https://www.adlinktech.com/en/ampere-based-solution
        
         | er4hn wrote:
         | This is a custom chip that they are making. I don't think that
         | they should be required to sell it, but if others find it
         | valuable you could expect to see hardware startups making their
         | own RISC-V AI chips as well that you could buy.
        
       | tmikaeld wrote:
       | Another AI chip made by TSMC, they make all of them?
        
         | jon-wood wrote:
         | TSMC manufacture (more or less) all of them. There's very few
         | companies in the world capable of manufacturing high
         | performance chips.
        
         | dragontamer wrote:
         | That's like complaining that all books are made by Penguin
         | Press or something, ignoring the effort individual authors
         | make.
         | 
         | Most of the value of chips is in their design, which is owned
         | by different entities. Manufacturing is important too (only
         | TSMC can make these advanced designs at scale and at lower
         | costs than the competition).
         | 
         | The question I have is if Cobalt has any innovations in its
         | design, or if its just bog-standard ARM Neoverse cores. Its not
         | too big of a deal to download ARM's latest designs and slap
         | them into... erm... your designs. But hopefully Microsoft added
         | value somewhere along the road (The Uncore remains important:
         | cache sizes, communications, and the like).
        
           | adrian_b wrote:
           | Microsoft claims that Cobalt has a much lower power
           | consumption than any other Arm CPUs that they have used.
           | 
           | Presumably this means that Cobalt has a much lower power
           | consumption than the current Ampere CPUs used by Azure.
           | 
           | Most of the power consumption reduction for a given
           | performance may have come from using a more recent TSMC
           | process, together with a more recent Arm Neoverse core, but
           | perhaps there might be also some other innovation in the MS
           | design.
        
           | solardev wrote:
           | It's not a matter of giving due credit, but supply
           | constraints. Books aren't limited by the availability of
           | printing presses, are they? (maybe they are and I just didn't
           | know?)
           | 
           | But if TSMC is the only company that can do this, they're a
           | bottleneck for the entire world. Not to mention a strategic
           | and geopolitical risk for the West.
           | 
           | It's be nice if some domestic companies invested in fabs
           | again...
        
       | lnsru wrote:
       | I wish I could work in some team designing these chips. Maia is
       | probably my dream product to work on. Super new, super cool and
       | one of it's kind.
        
         | geodel wrote:
         | I mean that looks cool and exciting if it is really small
         | colocated team if one is lead engineer or director of large
         | team of engineers so that they can learn do things that
         | interest them and assign boring/routine work to individuals.
         | Otherwise it would be just another job where people work on
         | assigned JIRA stories and go home in evening.
        
           | buildbot wrote:
           | Well, azure devops in this case ;)
        
         | aleph_minus_one wrote:
         | > I wish I could work in some team designing these chips. Maia
         | is probably my dream product to work on. Super new, super cool
         | and one of it's kind.
         | 
         | You likely became bewitched by their glamorous marketing side.
         | I'd bet that the real work that the team does is very similar
         | to the work that basically _every_ ASIC design team does.
        
           | asdfman123 wrote:
           | Think of all the requirements meetings!
        
         | hulitu wrote:
         | > Maia is probably my dream product to work on.
         | 
         | I bet you haven't used any Microsoft product before. /s
        
       | neilv wrote:
       | Initially looks pretty, but will mess you up:
       | https://en.wikipedia.org/wiki/Cobalt#Health_issues
        
       | opcode84 wrote:
       | The Microsoft chip has MX data types:
       | https://news.ycombinator.com/item?id=37930663
       | 
       | https://arxiv.org/abs/2310.10537
        
         | buildbot wrote:
         | Yes! Great catch :)
        
       | xnx wrote:
       | Chip manufacturers (including Nvidia) really missed where the
       | market was going if customers like Microsoft, Amazon, etc. feel
       | the need to make their own chips.
        
         | adgjlsfhk1 wrote:
         | I think they got the direction right but the price wrong. They
         | are used to dealing with super-computers as the main server
         | clients who aren't big enough to fight back if the prices creep
         | to high.
        
         | airstrike wrote:
         | Microsoft, Amazon, etc. feel the need to make their own chips
         | so that they don't let NVIDIA take all the profits, not because
         | they think NVIDIA is incompetent
        
         | vb-8448 wrote:
         | My guess is that it's more about the wish of cloud vendors to
         | control everything from the hw to sw: it's called vertical
         | integration, and it's common in a lot of businesses.
         | 
         | It makes a lot of sense from the point of view of cloud giants.
        
         | abraae wrote:
         | Or cloud vendors have decided that at their scale owning their
         | own chips represents a valuable differentiation opportunity,
         | and they don't think of them as commodities.
        
         | User3456335 wrote:
         | The demand for chips has increased so much that it's profitable
         | for these customers to start producing their own chips. This
         | doesn't mean Nvidia's chips are bad or that they missed
         | anything.
        
         | dragontamer wrote:
         | Alternatively: ARM has just made Neoverse designs that easy to
         | use that no one feels like going through another middleman
         | anymore.
        
         | lotsofpulp wrote:
         | As far as I understand, Nvidia does not manufacture chips, they
         | only design them and create the software.
        
         | justapassenger wrote:
         | It's just part of the cycle. For new paradigm, you have
         | companies jumping to build their own things. Later it unifies
         | and you get few main leaders.
         | 
         | It wasn't that long ago that computer manufacturers would build
         | their own chips.
        
         | solardev wrote:
         | Nvidia rode a gaming high from RTX straight into a crypto high
         | and then straight into the AI high. Their products just print
         | money right now and nobody else is close yet. They can always
         | lower prices later, but for now they're getting filthy rich...
        
       | MR4D wrote:
       | Every day we get closer and closer to the hypothetical "five
       | computers". [0] Software was first, and now every one of them
       | have hardware too.
       | 
       | Amazon
       | 
       | Apple
       | 
       | Goole
       | 
       | Microsoft
       | 
       | [0] - https://engines.egr.uh.edu/episode/1059
        
         | asdfman123 wrote:
         | Five mainframes, maybe. I've got at least five terminals for
         | those mainframes attached to, or inside of, my TV alone.
        
       | tibbydudeza wrote:
       | Does it run Linux or the Windows for Azure (definitely not a run
       | of the mill Windows for DataCenter).
        
       | kmod wrote:
       | I thought an interesting point was the liquid cooling -- unclear
       | how important this is to them, but I'm guessing it means that
       | they designed it with a TDP that requires liquid cooling.
       | 
       | This (wanting higher density) is the opposite of the trade-off
       | that I was expecting. In my (limited and out of date) experience,
       | power was the limiting factor before space, and I believe AI
       | racks have very high power draws already.
       | 
       | I would have guessed this would be because larger nodes would be
       | better for AIs tight communication patterns, but they
       | specifically call out datacenter space as the constraint. Curious
       | if anyone knows more about this
        
       | asdfman123 wrote:
       | What would be even more impressive would be making the chip in
       | the US so we weren't so completely reliant on TMSC.
        
       | ThinkBeat wrote:
       | I am a little confused on all this. I thought conventional wisdom
       | was that creating "super good/ top performant " computer
       | processors was hard, and unless you could produce them at scale
       | prohibitable expensive.
       | 
       | Is this done as a bridge until/if Nvidia is able to deliver their
       | processors fast enough?
       | 
       | I would think that competitors to Nvidea would have serious
       | competitors on the market already if competing can be done by
       | Microsoft for whom producing hardware is not their main business
       | focus.
        
         | p1esk wrote:
         | It remains to be seen how well these chips will compete with
         | Nvidia.
        
         | s0rce wrote:
         | Presumably Microsoft won't be making the chips, I guess they
         | will also be made by TSMC.
        
         | thetrb wrote:
         | Why would Microsoft not produce them at scale? With Azure and
         | internal use cases I'm sure they're at a good scale.
        
       | karmasimida wrote:
       | AI, or creation of AI, will very soon leave the hands of common
       | folks at this rate. The cost will soon unattainable.
        
         | er4hn wrote:
         | This is likely to be cyclical though. Fast cars[1], large
         | amounts of processing power[2], access to cryptographic
         | algorithms, are all things that started out expensive to be on
         | the cutting edge, some still are expensive on the cutting edge,
         | but then became more affordable for the consumer over time. AI
         | has already had explorations on training models with limited
         | resources. It's feasible, just has tradeoffs that will
         | hopefully get better over time.
         | 
         | [1] A 1970 Corvette 427 has a 0 - 60 mph of 5.3 seconds (src:
         | https://www.caranddriver.com/features/g15379023/the-
         | chevrole...) and cost around $44k inflation adjusted dollars.
         | You can buy a 2008 Nissasn 350Z Enthusiast that will do it in
         | 5.2 (src: https://www.zeroto60times.com/vehicle-
         | make/nissan-0-60-mph-t...?) for around $13k today.
         | 
         | [2] I'm too lazy to calculate relative cost / cycle in old
         | warehouse computers vs phones but it's gotten _better_.
        
       | mackid wrote:
       | "People who are really serious about software should make their
       | own hardware." - Alan Kay
        
       ___________________________________________________________________
       (page generated 2023-11-15 23:00 UTC)