Live data from Hacker News

Meta’s Hyperscale Infrastructure: Overview and Insights

cacm.acm.org

41–50 of 182 posts

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#41
post #9
post #7

Earlier quoted context omitted.

I thought the same thing, but then realize that Threads was a huge flop. Is it really a skill to very quickly release a dud app? I don't know the answer to that. Bypassing bureaucracy seems like heaven, but it feels like it also bypassed the product folks entirely.

What is threads lacking from a product perspective?

I guess its only real USP is "automatically import your Instagram friends", except that doesn't really work properly because only a fraction of people on Instagram seem to be interested in Threads.

Its fediverse integration stuff isn't panning out because nobody in the fediverse is stupid enough to let them federate. The single thread per message instead of hashtags thing doesn't seem to add a lot either.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#42
post #13

Not a word about Thrift, perhaps it was too low level for an infra overview, but I would have expected it to make some technical impact from a global perspective.

I wouldn't be surprised if they have switched to GRPC for improved performance. They mentioned that RPC libraries are centrally maintained in their monorepo; a migration from Thrift to GRPC might have taken less than 6 months.

I haven't been there a while, but that is extremely unlikely.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#43

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.

Precisely, for all the talk of efficiency last few years, how do we even begin to measure the total waste of effort and energy of so many smart people that this was? All that effort and stress for effectively nothing, or perhaps even net negative effect.

how is it net negative effect? they have Hundreds of millions of ACTIVE users

even if it was spun off to be stand alone product without Instagram it would still be worth billions

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#44
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…

Sounds like a horrific place to work. Imagine the pressure

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#45
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…

Sounds like a horrific place to work. Imagine the pressure

I think the joy of working at this scale makes up for that for a lot of people. There are plenty of low-stress, low-impact companies that someone that's bigtech-approved can go to

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#46
post #32

At least half their gak is due to them NOT moving quickly and NOT wanting to break things. IIRC, graphql is a means of papering over a bunch of legacy APIs. They removed foreign keys from mysql using it as a column store db, a vestige of the original LAMP stack still on PHP. I don't think Meta infrastructural choices are applicable to most folk. What does serverless land your average dev? A high AWS bill. Elastic man…

> Facebook creates tech to mitigate the pathologies their past continuously present.

Remember when they hacked a running Android Dalvik machine because their organizational constraints were such that they could never remove code or delete unused classes?

https://archive.is/nIPlg

https://engineering.fb.com/2013/03/04/android/under-the-hood...

Facebook seems like a place where they do amazing engineering to temporarily stave off the disastrous consequences of their previous feat of amazing engineering.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#47
post #20

Earlier quoted context omitted.

How many of those are bots?

How many users on X (Twitter) or Bluesky are bots? It's reasonable to assume the percentages are the same, given the lack of public information for the major text-based broadcast social networks. X is estimated to have 250M daily active users. Mark Zuckerberg recently stated that Threads' DAUs were 100M. Threads achieving a bit under half the size in such a short time is impressive, especially since Threads still lac…

> It's reasonable to assume the percentages are the same

It's reasonable to assume they're worse. Bluesky doesn't have Facebook's network or surveillance apparatus. Neither does Twitter, except it's a higher-value target than Bluesky and Threads combined.

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#48
post #14
post #3

Very interesting, in particular the explicit comparisons with hyperscalers. I almost wonder if this is preparation for them launching their own public cloud. Anyone from Meta care to comment?

I left before they were Meta, and maybe things have changed, but I don't think they have any intention of being a public cloud. Yes, they've got a lot of similar services as a public cloud, but there's a lot of opinionated choices that make sense for them that I think would be hard to convince customers to accept. Their infrastructure is cloudy, but it's built around mostly a single customer and assumes the infrastru…

> a lot of opinionated choices that make sense for them that I think would be hard to convince customers to accept

I get this vibe whenever I use the AWS or GCS dashboards, yet here we are!

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#49
post #7
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…

I thought the same thing, but then realize that Threads was a huge flop. Is it really a skill to very quickly release a dud app? I don't know the answer to that. Bypassing bureaucracy seems like heaven, but it feels like it also bypassed the product folks entirely.

It's certainly a skill to launch quickly at that scale. There are plenty of bureaucratically managed slowly launched duds too

I hate meta with a passion, but I don't deny they have some great infrastructure and engineers to enable the bad things they do to the world

Re: Meta’s Hyperscale Infrastructure: Overview and Insights

#50
post #7

Earlier quoted context omitted.

I thought the same thing, but then realize that Threads was a huge flop. Is it really a skill to very quickly release a dud app? I don't know the answer to that. Bypassing bureaucracy seems like heaven, but it feels like it also bypassed the product folks entirely.

What? It has more than 300M users.

I think most of those are Instagram shoving it in your face. Yeah I'm a "Threads user", but only because of the inline feed in Instagram. I'm annoyed when there is a notification blip but it turns out to be Threads spam.
Post reply on HN