A

Ahrefs Engineering Blog

tech.ahrefs.com · Company blog · on HN since 2017

Ahrefs' engineering blog covers running its own infrastructure for a web crawler and search index, along with OCaml, ReasonML, and hardware stats.

What HN saysHN links it for the post on saving $400M by avoiding the cloud, with commenters noting regular hosting is far cheaper, and the Skylake bug story.
32HN people shared it
11linked it in comments
22stories submitted
205story points
3 months agolast shared
2017people per year2026

Most-shared posts

  1. 1
    Ahrefs Saved US$400M in 3 Years by Not Going to the Cloud
    14 people · 19 points · first shared 2023 · 1 comments on HN
  2. 2
    Skylake bug: a detective story (2017)
    5 people · 1 points · first shared 2017 · discussion on HN
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
    Building Ahrefs Codebase with Melange
    1 people · 2 points · first shared 2021 · discussion on HN
  10. 10
  11. 11
    Ahrefs saved $400M in 3 years by not going to the cloud
    1 people · 2 points · first shared 2023 · 1 comments on HN
  12. 12
    How to get started with OCaml in 2022
    1 people · 2 points · first shared 2022 · discussion on HN
  13. 13
    One and a half years of ReasonML in production at Ahrefs
    1 people · 2 points · first shared 2020 · discussion on HN
  14. 14
    Profiling Dune Builds
    1 people · 1 points · first shared 2024 · discussion on HN
  15. 15

What HN people wrote when they linked it

kkfx · 2024-06-16
I have an European alternative, my own iron in my own home, it's European since I'm an EU citizen living in the EU... The alternative to the modern computing is the classic iron ownership instead of living on the shoulder of giants, otherwise there is no difference between a US company and an EU one selling the same services. They are both third party whose job is making money for their shareholders. "But.... HW cost money!", yes, because someone else computers are notoriously free... https://tech.ahrefs.com/how-ahrefs-gets-a-billion-dollar-worth-infrast… and of course owned hw is not a asset with an eventual resale value and a vast series of possibilities if a startup idea fail... Oh, yes, bandwidth might be tricky but co-location until a company is large enough to pay a personal proper link it's still an option, on their own hw. Dear devs startuppers where the recliner is the most valuable owned asset you have, it's about time to recalculate your economy.......
alfo_ · 2017-11-01
Ahrefs | OCaml developer / DevOps | Singapore + San Francisco, CA | REMOTE ok, VISA ok (SG only) Ahrefs runs an internet-scale bot that crawls the whole Web 24/7, storing huge volumes of information to be indexed and structured in a timely fashion. Backend system is powered by a custom petabyte-scale distributed key-value storage to accommodate all that data coming in at high speed. The storage system is implemented in OCaml with thin performance-critical low-level part implemented in C++ and Rust. On top of that Ahrefs is building various analytical services for end-users. We are a small team and strongly believe in better technology leading to better solutions for real-world problems. We worship functional languages and static typing, extensively employ code generation and meta-programming, value code clarity and predictability, and are constantly seeking to automate repetitive tasks and eliminate boilerplate, guided by DRY and following KISS. If there is any new technology that will make our life easier - no doubt, we'll give it a try. We rely heavily on opensource code (as the only viable way to build maintainable system) and contribute back [1]. Occasionally we track down CPU bugs [2]. We are looking for a backend developer [3] with a deep understanding of networks, distributed systems, OS fundamentals and taste for simple and efficient architectural designs. Our backend is implemented in OCaml (and some C++), as such proficiency in OCaml is very much appreciated,…
conradklnspl · 2024-08-28
While I agree that for user-facing APIs, serverless can both help handling load spikes and keep costs low, how would serverless help with spidering and indexing? Spidering and indexing are processes that would most likely run continuously for any search engine like Google or Kagi. There is always data to update and new web pages being created. Thus, they would benefit from dedicated servers from the get go for costs. Ahrefs posted an interesting article on this topic: https://tech.ahrefs.com/how-ahrefs-saved-us-400m-in-3-years-by-not-goi… Then again, you could use serverless on your own dedicated servers to benefit from the advantages of both.
lompad · 2024-06-20
>> somewhat It's a factor of 5 to 20 [1] in many cases. Regular hosting is unbelievably cheap, especially in high traffic environments. [1] https://tech.ahrefs.com/how-ahrefs-saved-us-400m-in-3-years-by-not-goi…
rwmj · 2024-02-23
Reminded me of the Intel Skylake bug found by the OCaml compiler developers: https://tech.ahrefs.com/skylake-bug-a-detective-story-ab1ad2beddcd
mobilio · 2023-06-10
rippercushions · 2023-03-13
@dang: the original blog post by Ahrefs is much more informative and detailed. https://tech.ahrefs.com/how-ahrefs-saved-us-400m-in-3-years-by-not-goi…
ThePowerOfFuet · 2023-03-11
CalChris · 2021-11-08
Debian announcement https://lists.debian.org/debian-devel/2017/06/msg00308.html Ahrefs writeup https://tech.ahrefs.com/skylake-bug-a-detective-story-ab1ad2beddcd The Intel spec update still labels SKL150 as No Fix but there is a microcode update available. Dunno exactly what to make of that distinction. https://www.intel.com/content/www/us/en/processors/core/desktop-6th-ge… Can an x86 program detect whether this update has been applied? Can a Linux process set a DONT_HYPERTHREAD_ME_BRO bit?

Keyboard shortcuts

/Search
j kNext / previous blog
EnterOpen selected blog
vVisit the blog site
fFollow / unfollow
lSwitch grid / list view
dSwitch dark mode
g hHome
g pTop posts
g tTopics
g fFollowing
g lLiked
g yHistory
g qFAQ
EscBack / close