Live data from Hacker News

TPUs vs. GPUs and why Google is positioned to win AI race in the long term

uncoveralpha.com

211–220 of 328 posts

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#211
post #194

Earlier quoted context omitted.

Nvidia doesn't have the software stack to do a TPU. They could make a systolic array TPU and software, perhaps. But it would mean abandoning 18 years of CUDA. The top post right now is talking about TPU's colossal advantage in scaling & throughput. Ironwood is massively bigger & faster than what Nvidia is shooting for, already. And that's a huge advantage. But imo that is a replicateable win. Throw gobs more at netwo…

> They could make a systolic array TPU and software, perhaps. But it would mean abandoning 18 years of CUDA. Tensor cores are specialized and have CUDA support.

Tensor cores can help a lot for matrix maths, sure, definitely. They made a big splash in 2017 & have been essential. https://developer.nvidia.com/blog/programming-tensor-cores-c...

But it's still something grafted onto the existing architecture, of many grids with many blocks with many warps, and lots and lots of coordination and passing intermediary results around. It's only a 4x4x4 unit, afaik. There's still a lot of main memory being used to combine data, a lot of orchestration among the different warps and blocks and grids, to get big matrices crunched.

The systolic array is designed to allow much more fire and forget operations. It's inputs are 128 x 128 and each cell is its own compute node basically, shuffling data through and across (but not transitting a far off memory).

TPU architecture has plenty of limitations. It's not great at everything. But if you can design work to flow from cell to neighboring cell, you can crunch very sizable chunks of data with amazing data locality. The efficiency there is unparalleled.

Nvidia would need a radical change of their architecture to get anything like the massive data locality wins a systolic array can do. It would come with massively more constraints too.

Would love if anyone else has recommended reading. I have this piece earmarked. https://henryhmko.github.io/posts/tpu/tpu.html https://news.ycombinator.com/item?id=44342977

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#213

> It is also important to note that, until recently, the GenAI industry’s focus has largely been on training workloads. In training workloads, CUDA is very important, but when it comes to inference, even reasoning inference, CUDA is not that important, so the chances of expanding the TPU footprint in inference are much higher than those in training (although TPUs do really well in training as well – Gemini 3 the prim…

This is a very important point - the market for training chips might be a bubble, but the market for inference is much, much larger. At some point we might have good enough models and the need for new frontier models will cool down. The big power-hungry datacenters we are seeing are mostly geared towards training, while inference-only systems are much simpler and power efficient. A real shame, BTW, all that silicon d…

it's already the case that people are eeking out most further gains through layering "reasoning" on top of what existing models can do - in other words, using massive amounts of inference to substitute for increases model performance. Whereever things plateau I expect this will still be the case - so inference ultimately will always be the end game market.

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#214

Will Google sell TPUs that can be plugged into stock hardware, or custom hardware with lots of TPUs? Our customers want all their video processing to happen on site, and don't want their video or other data to touch the cloud, so they're not happy about renting cloud TPUs or GPUs. Also it would be nice to have smart cameras with built-in TPUs.

Why don't your customers trust Google Cloud?

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#215
I always enjoy being wrong and I was very wrong in my predictions about Google : I thought they should theoretically win, but I was also very confident they couldn't possibly turn their execution ship around to actually pull together a coherent competitor to OpenAI. But they do seem to have done that and it's very impressive. If they do continue to execute, I can't see anybody stopping them dominating and I would be bearish on nearly every other player catching them.

The biggest problem though is trust, and I'm still holding back from letting anyone under my authority in my org use Gemini because of the lack of any clear or reasonable statement or guidelines on how they use your data. I think it won't matter in the end if they execute their way to domination - but it's going to give everyone else a chance at least for a while.

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#216
post #176
post #41

Google's real moat isn't the TPU silicon itself—it's not about cooling, individual performance, or hyper-specialization—but rather the massive parallel scale enabled by their OCS interconnects. To quote The Next Platform: "An Ironwood cluster linked with Google’s absolutely unique optical circuit switch interconnect can bring to bear 9,216 Ironwood TPUs with a combined 1.77 PB of HBM memory... This makes a rackscale…

No, not at all. If this were true Google would be killing it in MLPerf benchmarks, but they are not. It’s better to have a faster, smaller network for model parallelism and a larger, slower one for data parallelism than a very large, but slower, network for everything. This is why NVIDIA wins.

I mean, Google just isn't participating it seems?

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#217
post #215

I always enjoy being wrong and I was very wrong in my predictions about Google : I thought they should theoretically win, but I was also very confident they couldn't possibly turn their execution ship around to actually pull together a coherent competitor to OpenAI. But they do seem to have done that and it's very impressive. If they do continue to execute, I can't see anybody stopping them dominating and I would be…

The LLM provider I trust the most right now is AWS. Anybody else seems to have very conflicted purposes when it comes to sending them my data and interactions.

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#219
post #2

A question I don't see addressed in all these articles: what prevents Nvidia from doing the same thing and iterating on their more general-purpose GPU towards a more focused TPU-like chip as well, if that turns out to be what the market really wants.

It’s not binary. It’s not existential. What’s at stake for Nvidia is its HUGE profit margins. 5 years from now, Nvidia could be selling 100x as many chips. But its market cap could be a fraction of what it is now if competition is so intense that its making 5% profit margin instead of 90%.

More like 900% right now.

Re: TPUs vs. GPUs and why Google is positioned to win AI race in the long term

#220

In my 20+ years of following NVIDIA, I have learned to never bet against them long-term. I actually do not know exactly why they continually win, but they do. The main issue they have a 3-4 year gap between wanting a new design pivot and realizing it (silicon has a long "pipeline"), it can seem that they may be missing a new trend or swerve in the demands of the market, it is often simply because there is this delay.

Turkeys bet on tomorrow 364 days of the year.
Post reply on HN