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

Building Meta's GenAI infrastructure

engineering.fb.com

Post reply on HN
Print view View original

301–310 of 314 posts

Re: Building Meta's GenAI infrastructure

#301

Post by nradov »

brookst wrote: post #242
boringg wrote: post #240

Earlier quoted context omitted.

Front loading costs to eventually extract rents on usage with one hell of a capital wall protecting the assets. Its easier to spin up a business for sure -- also easier to unwind it - there not as sticky as they used to be.

This is typically called a high fixed cost business, like airlines, hotels/apartments, SpaceX, etc. The dream may be barriers to entry that allow high margins (“rents” if you prefer the prejudicial), but all too often these huge capital costs bankrupt the company and lose money for investors (see: WeWork, Magic Leap). It is high risk, high return. Which seems fair.

Nothing in the WeWork business model is inherently capital intensive. Fundamentally they just take out long-term office leases at low rates, then sublease the space to short-term tenants at higher rates. They don't really own major assets and have no significant IP.

Re: Building Meta's GenAI infrastructure

#302

Post by herval »

Jensson wrote: post #278
herval wrote: post #248

Earlier quoted context omitted.

I’m not sure we went through the same dot-com era, but in my experience, it was extremely expensive to spin up anything. You’d have to run your own servers, buy your own T1 lines, develop with rudimentary cgi… it was a very expensive mess - just like AI today Which gives me hope that - like the web - hardware will catch up and stuff will become more and more accessible with time

> I’m not sure we went through the same dot-com era, but in my experience, it was extremely expensive to spin up anything. You’d have to run your own servers, buy your own T1 lines, develop with rudimentary cgi… it was a very expensive mess - just like AI today To make your own competing LLM today you need hundreds of millions of dollars, the "very expensive" of this is on a whole different level. You could afford th…

Only if you're creating a foundation model. The equivalent would be competing with a well-funded Amazon, back in 1999. You can compete in building LLM-powered products with much, much less money - less than a regular web app in 99

Re: Building Meta's GenAI infrastructure

#303

Post by herval »

anon373839 wrote: post #286
Jensson wrote: post #278

Earlier quoted context omitted.

> I’m not sure we went through the same dot-com era, but in my experience, it was extremely expensive to spin up anything. You’d have to run your own servers, buy your own T1 lines, develop with rudimentary cgi… it was a very expensive mess - just like AI today To make your own competing LLM today you need hundreds of millions of dollars, the "very expensive" of this is on a whole different level. You could afford th…

I think the foundation models are a commodity, anyway. The bulk of the economic value, as usual, will be realized at the application layer. Building apps that use LLMs, including fine-tuning them for particular purposes, is well within reach even of indie/solo devs. That’s why Sam Altman makes so much noise about “safety” - OpenAI would really like a government-backed monopoly position so they can charge higher rents…

I think openai/anthropic/etc are banking on foundation models being the equivalent of the "datacenters" or AWS-equivalents of AI - there'll be PaaSes (eg replicate), and most businesses will just pay the "rent"

Re: Building Meta's GenAI infrastructure

#304

Post by whatshisface »

woooooo wrote: post #291
whatshisface wrote: post #241

Earlier quoted context omitted.

If the government can stay back far enough that more than one AI company can train their models, it will end up working like steel mills - barely enough profit to pay the massive cost of capital due to competition. If the government regulates the industry into a monopoly, all bets are off. Their investors are going to push hard for shutting the door behind them so watch out. The only question is - what tactic? I don'…

Funny example, US Steel was a textbook case for a monopoly achieved privately because it wasn't regulated against.

They were on top for a while, but later fell behind because they didn't invest. There were heavy tarrifs in place to "protect" the monopoly from foreign competition.

Re: Building Meta's GenAI infrastructure

#305

Post by virtuallynathan »

jedberg wrote: post #300
virtuallynathan wrote: post #296

Earlier quoted context omitted.

Yep, I am aware, I have a square footage multiplier for their multi-story buildings.

But how can you know how many floors they have? And where are you getting the list of buildings from? And what makes you think your list is complete? Also how do you know their efficiency? Google might have less space but also a way to pack twice as much compute in the same place. Like I said, this is impossible to know without a lot of insider information from a lot of companies.

Well, they tell you for one: https://datacenterpost.com/scaling-up-google-building-four-s... We also have Google Street view, etc.

Of course, it's all estimates. You can get fancier and count generators and transformers and stuff too.

Re: Building Meta's GenAI infrastructure

#306

Post by virtuallynathan »

virtuallynathan wrote: post #182
virtuallynathan wrote: post #126

Earlier quoted context omitted.

Facebook has more datacenter space and power than Amazon, Google, and Microsoft -- possibly more than Amazon and Microsoft combined...

To date, facebook has built, or is building, 47,100,000 sq ft of space, totaling nearly $24bn in investment. Based on available/disclosed power numbers and extrapolating per sqft, I get something like 4770MW. Last I updated my spreadsheet in 2019, Google had $17bn in investments across their datacenters, totaling 13,260,000 sq ft of datacenter space. Additional buildings have been built since then, but not to the sca…

I updated my map for AWS in Northern Virginia -- came up with 74 buildings (another source says 76, so i'll call it directionally correct). If I scale my sq ft by ~5% to account for missing buildings, we get 11,500,000sq ft in the northern virginia area for AWS.

I'll finish my other maps and share them later...

Re: Building Meta's GenAI infrastructure

#307

Post by woooooo »

whatshisface wrote: post #304
woooooo wrote: post #291

Earlier quoted context omitted.

Funny example, US Steel was a textbook case for a monopoly achieved privately because it wasn't regulated against.

They were on top for a while, but later fell behind because they didn't invest. There were heavy tarrifs in place to "protect" the monopoly from foreign competition.

All true, but the initial agglomeration was 100% private

Re: Building Meta's GenAI infrastructure

#308

Post by whatshisface »

woooooo wrote: post #307
whatshisface wrote: post #304

Earlier quoted context omitted.

They were on top for a while, but later fell behind because they didn't invest. There were heavy tarrifs in place to "protect" the monopoly from foreign competition.

All true, but the initial agglomeration was 100% private

If privately arising monopolies could only be kept from buying out their regulators, they'd privately break down before they became too odious... for example Google, which for years was the only remotely good search engine, is now merely one of the better search engines. If there had been a "department of consumer information safety," staffed by the best industry professionals status can buy, that might have not happened.

Re: Building Meta's GenAI infrastructure

#309

Post by althea_tx »

motoxpro wrote: post #31
renegade-otter wrote: post #25

Earlier quoted context omitted.

It will be ironic if Meta sinks all this money into the new trend and finds out later that it has been a huge boondoggle, just as publishers followed Facebook's "guidance" on video being the future, subsequently gutting the talent pool and investing into video production and staff - only to find out it was all a total waste.

It already paid off. When the world moved from determinisic to probablistic ad modeling. That's why their numbers are so good right now compared to every other advertiser

Can you explain more about this?

Re: Building Meta's GenAI infrastructure

#310

Post by edmundsauto »

verticalscaler wrote: post #113
oblio wrote: post #108

Earlier quoted context omitted.

The thing is, this could be considered basic research, right? Basic research IS setting money on fire until (and if) that basic research turns into TCP/IP, Ethernet and the Internet.

I wish. Funnily enough Arpanet and all that Xerox stuff were like Where as I think this more appropriately can be considered the meta PR budget. They simply can't not spend it, would look bad for Wall Street. Have to keep up with the herd.

This suffers from hindsight bias, at the time it was impossible to know if Arpanet or flying cars was the path forward. A better comparison would be the total sum of investment : payoff ratio, and is not something we can see from where we are now. Only in the future does it make sense to evaluate the success of something. Unfortunately, comparison between eras is difficult to do fairly because conditions are so different between now and Xerox.
Post reply on HN