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

Meta’s Hyperscale Infrastructure: Overview and Insights

cacm.acm.org

Post reply on HN
Print view View original

101–110 of 182 posts

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#101

Post by ramblemonkey »

yuliyp wrote: post #2

I find it interesting how they describe the PHP web front end as a "serverless" or "function as a service" architecture. I guess it's a matter of perspective. It's a service that has a monolithic codebase with lots of endpoints deployed to it. I guess from the perspective of the maintainer of one of those endpoints it's "serverless" but that abstraction (like all abstractions) has leaks: the teams responsible for the…

There’s only one entry point for all requests. It’s php (well, Hack) all the way down from there. So all the routing is readable as php code, request handling etc.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#102

Post by factfindingisfn »

Sponsored by Brawndo - its what plants crave

Seriously though, this is painful to watch. Reminds me of those articles/videos where people have to have their jobs "invest" or software/hardware partners in that website and all of a sudden this person is a "Person of Interest".

Meh

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#104

Post by NBJack »

vasco wrote: post #4

> Moreover, once it [Threads] was developed, the infrastructure teams were given only two day’s notice to prepare for its production launch. Most large organizations would take longer than two days just to draft a project plan involving dozens of interdependent teams, let alone execute it. At Meta, however, we quickly established war rooms across distributed sites, bringing together both infrastructure and product te…

Given the looming threat of sudden job loss at Meta for the past few years now, I'm not surprised. Fear can motivate, if temporarily.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#105

Post by NBJack »

linkregister wrote: post #30
sn0wleppard wrote: post #27

Earlier quoted context omitted.

And how many are Instagram users who created a Threads account after being prompted but never actually use it?

Daily Active Users (DAUs) and Monthly Active User counts represent the amount of users that perform activity on the application daily and monthly, respectively.

There is no industry standard for DAUs or MAUs. Take it all with a hefty grain of salt.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#106

Post by NBJack »

yuliyp wrote: post #2

I find it interesting how they describe the PHP web front end as a "serverless" or "function as a service" architecture. I guess it's a matter of perspective. It's a service that has a monolithic codebase with lots of endpoints deployed to it. I guess from the perspective of the maintainer of one of those endpoints it's "serverless" but that abstraction (like all abstractions) has leaks: the teams responsible for the…

It's serverless only in that there's a whole different organization at Meta handling the (monolith) hosting on actual hardware.

By that definition, you could argue Amazon's detail page is "serverless".

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#108

Post by njovin »

philg_jr wrote: post #107

Amazing. All of this truly wild, impressive technology and some of the absolute best engineers in the world, just to shove more ads in people's eyeballs. Sigh.

Hey now let's not downplay what a wonderful vehicle it is for state-sponsored political propaganda and misinformation.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#109

Post by NBJack »

nk8620 wrote: post #76

> Moreover, for non-AI compute workloads, we offer only a single server type, equipped with one CPU and the same amount of DRAM (previously 64GB, now 256GB). I'm reading something like this for the first time. Is this common across industry or only typical to Meta? In contrast, we use multiple instances for sub-components of our ML training pipeline.

Not to my knowledge, no. If it's a virtualized server, maybe this works, but to my knowledge most companies try to maximize their hardware usage through a variety of slices/sizes based on underlying resources. I am not a datacenter expert however.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#110

Post by datadrivenangel »

Retr0id wrote: post #52
Simon_O_Rourke wrote: post #28

Earlier quoted context omitted.

They might have launched it quickly, but nobody (relatively speaking) gave a tinkers damn about the end product. All form and no function.

Threads feels more like a new feature in the facebook meta-app (heh) than a new app in its own right, especially with how it has to be bound to an instagram profile. "our new feature reached 100m users in 5 days" sounds a little less impressive (especially given meta has multiple billions of users to start with).

And how easy it was to start the threads onboarding flow inside FB/Insta... looked like a new feature.
Post reply on HN