Live data from Hacker News

Elixir saves Pinterest $2M a year in server costs

paraxial.io

341–350 of 481 posts

Re: Elixir saves Pinterest $2M a year in server costs

#341

Guys, Pinterest hasn't used Elixir at all for more than half a decade. I think there was a single rate limiting service that used it and I have no idea if that was swapped out or not.

This is incorrect. We continue to use it in a large rule engine for spam filtering.

Re: Elixir saves Pinterest $2M a year in server costs

#342
post #223

Earlier quoted context omitted.

Sure, but the fastest Elixir-based entry on that list has half the score of uvicorn (Python), so not really sure what your point is here w.r.t. OP.

Reading higher in the thread, some Elixir folks are saying that the techempower benchmarks used the wrong settings (debug mode, etc) for their Elixir benchmark.

Though that doesn’t change the crux of my question (rather it reinforces it): why invoke that benchmark in this context at all?

Re: Elixir saves Pinterest $2M a year in server costs

#343

Earlier quoted context omitted.

Obesity is not due to an eating disorder, it's due to sedentary lifestyles, pedestrian hostile neighborhoods, and market forces in the food industry.

IME it is usually mental health related.

Honestly, most of the people I know who have serious problems with obesity have some kind of metabolic disorder. Diabetic or pre-diabetic.

And I think this has become far more common than people will admit. We're burning out our pancreases through processed & high sugar, low-GI foods because the industrial food supply is poorly managed and under-regulated.

That plus cars.

Re: Elixir saves Pinterest $2M a year in server costs

#344
post #334

Earlier quoted context omitted.

this has been inconsistent over the years. maybe they 'fixed' it but it wasn't not the case before. they were extremely parasitical. you'd click the button and it would just take you to more pins

that only happens when the image was manually uploaded by a user (i.e from gallery). there's not much they can do about it. in that case you can scan it to find visual matches and hopefully one of them has metadata about the original source (but it's not perfect and you have to scroll a bit).

I'll take your word for it, you're welcome to it, but I'd rather it just didn't show up so high in search results.

Oh, and if it could stop recommending pro-anorexic content to my teenager, I'd like that to. It's banned on the LAN but I can't enforce that elsewhere.

It's a dumpster fire of SEO dark patterns and unmoderated shitty content and shouldn't be ranked so high in searches. Google should have routed around them years ago.

Re: Elixir saves Pinterest $2M a year in server costs

#345
>Bleacher Report, a top real time sports and media website, was hitting scaling problems due to their business success. With Elixir, they were able to 8x the average daily traffic load, going from 150 servers to only 5: The site was originally written on the Ruby on Rails framework but Bleacher Report reached the point where they could no longer scale it, according to Dave Marks, BR’s senior engineering director.

And they said they could do it with 2. That is anywhere from 30x to 75x. Although I think the current work on Fibre and Async could have reduced it by 10x to 20x as well. Rails just need to adopt it. It is unfortunate they didn't try to optimise it but instead went with Elixir.

Re: Elixir saves Pinterest $2M a year in server costs

#346
post #14

Does anyone has any followup information about the companies and people mentioned? Sadly last blog from Cory O'Daniel is from 2019 and Bleacher Report engineering blog post shows 2019 as the last post. Pinterest has 81 public repos on GitHub but only 3 with Elixir code base.

Elixir still gets limited use at Pinterest. I think there's only one significant system still actively developed in Elixir, though it's pretty non-trivial both in terms of cost and complexity.

Re: Elixir saves Pinterest $2M a year in server costs

#347

I use Go and it gives me a good sweet spot between cost and perf. I’m nowhere close to performance requirements of Pinterest. But my auth service runs on a single 1vcpu, 512mb ram, and 10gb ssd. I use leveldb and swap on the 10gb. I’ve benchmarked it to handle 8-9k rps while delivering 150ms max response time. Not bad for a few bucks a month.

And? Comparing apples and oranges?

I'm running seven Elixir/Phoenix apps (one of these is umbrella with 12 apps) on shared CPU cheap VPS with 1gb memory. I actually ran out of disk space recently than anything else so far.

Re: Elixir saves Pinterest $2M a year in server costs

#348

Guys, Pinterest hasn't used Elixir at all for more than half a decade. I think there was a single rate limiting service that used it and I have no idea if that was swapped out or not.

This is incorrect. We continue to use it in a large rule engine for spam filtering.

I was there and it was not being used in new projects. Maybe that spam filtering service has existed for that long. I would definitely not say that pinterest has fully bought into using elixir because of cost savings which is what the article seems to be insinuating.

Re: Elixir saves Pinterest $2M a year in server costs

#349
post #238
post #199

Earlier quoted context omitted.

I wrote the same kind of tool in both Elixir and Go, the performance difference was huge between the two, Elixir / Erlang is pretty slow overall.

The go runtime has similar capabilities as the BEAM runtime when it comes to concurrent workloads. Go has the benefit of being a typesafe compiled language which gives it speed benefits. But using either one of them instead of Java is probably going to be a huge win for most teams on concurrent workloads.

> The go runtime has similar capabilities as the BEAM runtime when it comes to concurrent workloads.

Only if you think that the BEAM is similar to being able to easily spawn a function on a separate thread and having channels.

Last I checked, goroutines had no mailboxes, supervisors, process monitoring, registry, and its scheduler has a much smaller scope and featureset than the BEAM's.

I swear it's obvious when people comment about the Erlang ecosystem without having really used it for anything.

Re: Elixir saves Pinterest $2M a year in server costs

#350
post #201
post #64

Earlier quoted context omitted.

The advantages of Elixir are not performance-related. There is a lot of focus on raw performance on web-related services, when in reality most of their running time is spent waiting for IO. If there are two things the BEAM excels at, is IO and turning almost any problem into half a dozen processes that are scheduled and run in parallel, if not geographically-distributed, with 1/50th the effort of any other language.…

Elixir does not do anything better at IO than other languages.

What's "other languages"?

IO lists, which are the foundation of anything that build a string incrementally, are managed with vectored IO syscalls out of the box (readv/writev), which you'd have to handle yourself in most other languages, or resort to allocating and endless memory copying which Erlang is able to avoid.

Post reply on HN