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

How Precision Time Protocol is being deployed at Meta

engineering.fb.com

Post reply on HN
Print view View original

41–50 of 104 posts

Re: How Precision Time Protocol is being deployed at Meta

#41

Post by NBJack »

bombcar wrote: post #34
CapmCrackaWaka wrote: post #6

Why do so many tech companies seem to be releasing “secret sauce” for free lately? I see a lot of posts lately detailing how inner production systems work at large companies, and while I’m grateful, I’m curious why the higher ups think it’s worthwhile to release this information.

It's an effective marketing campaign for "you should work here" - VERY effective. And lots of this stuff is NOT secret sauce, it's basic business building-blocks that they need. It's not the advertising formulas. I'm sure FAANG is very VERY happy that they can just run Linux everywhere and don't have to pay Sun or Microsoft a massive per-CPU fee for everything they do.

I think you'd normally be right, but Meta just laid off 10k+ people, and is currently in a company wide hiring freeze until at least Q1 of 2023. Much of the rest of FAANG is either doing one or both as well.

In Meta's case, they fired a lot of "boot campers" as well, some only a few days into their job and before they had a team. Some returning interns even had their offers rescinded.

Not to sound cynical, but this article feels more like "let us release this to show impact and keep our jobs." Or damage control for their engineering hiring image.

Re: How Precision Time Protocol is being deployed at Meta

#43

Post by loeg »

jtsiskin wrote: post #14
epgui wrote: post #2

This is incredible. If I knew I'd be working on things like this (as opposed to optimizing ad revenue or "user engagement" by A/B testing button copy), I'd possibly consider working at Meta.

At Meta you choose what team you want to work on during a 3 month “bootcamp” period, very different from Google, Amazon, etc. There’s enough teams solving hard distributed system problems that you’re guaranteed to never need to think about button copy

You can also be directly hired onto a team, skipping bootcamp (some roles).

Re: How Precision Time Protocol is being deployed at Meta

#44

Post by bombcar »

NBJack wrote: post #41
bombcar wrote: post #34

Earlier quoted context omitted.

It's an effective marketing campaign for "you should work here" - VERY effective. And lots of this stuff is NOT secret sauce, it's basic business building-blocks that they need. It's not the advertising formulas. I'm sure FAANG is very VERY happy that they can just run Linux everywhere and don't have to pay Sun or Microsoft a massive per-CPU fee for everything they do.

I think you'd normally be right, but Meta just laid off 10k+ people, and is currently in a company wide hiring freeze until at least Q1 of 2023. Much of the rest of FAANG is either doing one or both as well. In Meta's case, they fired a lot of "boot campers" as well, some only a few days into their job and before they had a team. Some returning interns even had their offers rescinded. Not to sound cynical, but this a…

"Positive press" is just preemptive damage control.

And anyone knows that a "company wide hiring freeze" is surprisingly thawable in cases where it needs to be.

Re: How Precision Time Protocol is being deployed at Meta

#45

Post by gpderetta »

jeffbee wrote: post #19

Great article, but I have to quibble about the casual ease with which they wave away the complexity of accessing these timestamps. The Linux APIs for accessing them is totally absurd. See for example the gRPC code that associates hardware (possibly) timestamps with messages, for tracing and other reasons. You have to re-arm the timestamp option before every send. And the whole concept of a timestamp for an ethernet f…

I don't see the code rearming it at every send nor it matches my experience. The code sets the sock option lazily, and only once . The option is needed to configure the driver and possibly the card. Then the timestamp option need to be passed to the send/recvmsg call to phisically retieve the send/recv timestamp.

Far for being absurd it seems a very straightforward extension of BSD socket API.

Re: How Precision Time Protocol is being deployed at Meta

#46

Post by pcthrowaway »

synaesthesisx wrote: post #37
zzixp wrote: post #16

I've been seeing a lot more of these Meta engineering articles on HN - I'm a fan! If you're writing them, keep up the great work!

Yes! I had no idea Meta actually invested in innovation/engineering like this. I always assumed most of their work was just in ad tech & "metaverse".

I mean, they've created a whole bunch of the tech required to support their scale (React, Jest, PyTorch)

Re: How Precision Time Protocol is being deployed at Meta

#47

Post by elcomet »

epgui wrote: post #35
capableweb wrote: post #33

Earlier quoted context omitted.

> Meta/FB for years and have never touched user-facing code or ads stuff Does that really matter in the end? If you're against writing code for ads/optimizing for quick dopamine hits, is it really so different to be writing the code that shows the ads vs the code that sets up the infrastructure to serve the ads? Or even infrastructure to support the developers who write the code for the ads? If you don't want to supp…

Yeah that's exactly my concern! I want to contribute to building cool things, and building a really cool component becomes much less inspiring if it's only used to steer nuclear missiles. (dramatic but not unreasonable example)

You could apply to work on the research infrastructure (for instance for AI)

Re: How Precision Time Protocol is being deployed at Meta

#48

Post by pcthrowaway »

NBJack wrote: post #41
bombcar wrote: post #34

Earlier quoted context omitted.

It's an effective marketing campaign for "you should work here" - VERY effective. And lots of this stuff is NOT secret sauce, it's basic business building-blocks that they need. It's not the advertising formulas. I'm sure FAANG is very VERY happy that they can just run Linux everywhere and don't have to pay Sun or Microsoft a massive per-CPU fee for everything they do.

I think you'd normally be right, but Meta just laid off 10k+ people, and is currently in a company wide hiring freeze until at least Q1 of 2023. Much of the rest of FAANG is either doing one or both as well. In Meta's case, they fired a lot of "boot campers" as well, some only a few days into their job and before they had a team. Some returning interns even had their offers rescinded. Not to sound cynical, but this a…

People are still getting hired at Facebook/Meta, they're probably just targeting specific people or hiring in cheaper countries now

Re: How Precision Time Protocol is being deployed at Meta

#49

Post by bsder »

CapmCrackaWaka wrote: post #6

Why do so many tech companies seem to be releasing “secret sauce” for free lately? I see a lot of posts lately detailing how inner production systems work at large companies, and while I’m grateful, I’m curious why the higher ups think it’s worthwhile to release this information.

Resume building for people to get their next job?

Lots of these companies opened up the ability for internal engineers to write tech blogs for the purposes of recruiting since the FAANGs were all in competition with one another. Presumably, the VPs haven't gotten around to closing the pipeline yet.

Re: How Precision Time Protocol is being deployed at Meta

#50

Post by pclmulqdq »

throw0101a wrote: post #7

So they state: > One could argue that we don’t really need PTP for that. NTP will do just fine. Well, we thought that too. But experiments we ran comparing our state-of-the-art NTP implementation and an early version of PTP showed a roughly 100x performance difference: While I'm not necessarily against more accuracy/precision, what problems specifically are experiencing? They do mention some use cases of course: > Th…

I have never seen NTP in a datacenter of reasonable size get below an error of 1e-4. PTP without custom hardware can easily get 3 orders of magnitude better.
Post reply on HN