AI Datacenters Were Built for GPUs. What Happens When You Remove the GPUs? | Hacker News Reader

AI Datacenters Were Built for GPUs. What Happens When You Remove the GPUs?

almartis.xyz

The whole website is genuinely unreadable / illegible with poor contrast.

This is the third time I've seen a website with this styling (serif, yellow and white on black). What's going on? Is it a template or some AI induced convergence?

Yeah, with only minimal guidance it's what you get out of claude. The colours and layout are pretty 'default'.

So, I do understand people using LLMs to do websites, they want to communicate some idea, and "typing HTML" isn't part of that, fine, use the LLM to slop it together, whatever.

But don't people review these things before they make it public? The website is borderline unreadable, how does this happen? Am I wrong for assuming people generate a website, review it and then deploy it? Do they only review the source, generate a website, asks Claude to review it, Claude says "Looks good" and the author just goes with it?

I'm struggling to understand why so many of the websites are so unreadable, when it's so easy to spot and fix these issues, it's like people are lazymaxxing nowadays, and not even in the fun "I'm a good developer because I'm lazy", just people being lazy-lazy.

I think it's Claude, yeah.

If I had to guess they used this skill https://impeccable.style

Default output of claude code. Another obvious example is https://trumprx.gov/, with the background beige that's kinda close to the Hacker News one (to my eyes at least)

For what ever reason reader mode appears to show different content

It does for me too, in Safari on iPhone.

Looks like a font rendering or retina issue

yes, they should increase the font size as a first step.

I can do "ctrl + +" to increase the font, but it's still serif and low contrast, so I have to do "ctrl + A". Or better yet - press "reader view" on firefox.

And the text appears to fade in on scroll, so you see literally nothing while scrolling down. Wonderful UX. It's like deliberately reintroducing a worse version of the checkerboard pattern the original iPhone used to display when the CPU couldn't handle the page rendering in time.

> For the past few decades, building a datacenter has been a well-understood, predictable exercise in utility engineering.

> In modern AI clusters, the network is no longer just infrastructure sitting beneath compute

It always make me smile when someone is presenting these kind of topologies as "New", "Modern A.I" or anything remotely "Revolutionary".

The HPC domain and any decent supercomputers have been doing RDMA networking centered around "all-to-all" and "all-reduce" operations for at least 3 fucking decades now.

They are the main reasons supercomputers are almost always constructed around stupidly complex Torus or Dragonfly network topologies.

MPI itself has these primitives defined from v1.

The only difference now is that it switch from "This niche thing 3 nerds were using for weather simulations" to "this cool thing any hyperscaler NEED to have for *A.I*"

> someone

in this context, that someone is an AI bot that spat out words.

>This allows us to flatten the physical datacenter into a GPU-free, non-blocking, 1-tier full mesh architecture built around high-density CPU nodes and 51.2Tb silicon switching fabric.

I wonder how many CPUs will be required to do the job of one Nvidia H200 GPU.

I never considered the implications and impacts on datacenters' architecture and organisation. It’s fascinating.

We're two data center networking engineers who've spent years designing and operating data center infrastructure for governments, telcos and banks in West Africa. This piece came out of our work on a new AI architecture based on associative memory rather than transformers. The GPU-free argument here is something we think about the next phase of AI networking. Happy to discuss further about it.

PS: Taking a look at our manifesto (https://almartis.xyz/) can help with more context.

Very interesting of course, but stuff like this just needs a demo not a book. Can be super simple, but it needs to be demonstrated somehow.

Yes, read that. What these people are talking about seems to replacing training of NNs by something else entirely. The big question is, does that work? At all?

It's premature to discuss network architecture until that basic question is answered.

I'm maybe 10% of the way in but I find I'm increasingly skeptical. If the basic building block dates back to the 1970s haven't other people tried this by now? If not, isn't the first order of business to throw together a prototype that solves MINST or one of the many other small datasets floating around out there as a proof of concept?

So unfortunately I'm inclined to assume this is empty conjecture shat out by an LLM. Because who would write something up in this much detail rather than typing `import numpy as ...` and going to town?

I'll also note that the document has all the usual crank signs. Lots of grand visions, hypotheses, and expounding at an overly high level on how various things work with hardly anything concrete.

Datacenters are being built for AI. What happens when you remove the AI workload?

Don't get me wrong. I don't mind when some tech bros burn billions of venture capital & nothing much (?) comes out of it.

But those datacenters embody a lot of resources. Raw materials, complex/resource heavy manufacturing processes for IC's, servers, networking gear, etc etc.

I sure hope that doesn't go to waste when the AI bubble pops. Datacenter stuffed with AI optimized hardware any good for general engineering? Science projects? Weather prediction? Web hosting? ...??

Well there's always predictive policing, fully automated facial recognition, chat control surveillance systems, nation-scale fingerprinting, and location tracking via inference of arbitrary signals.

Rendering commercials, probably.

Reply on news.ycombinator.com