Top all-time · #9568
Tessil
Personal blog · tessil.github.ioprogrammingsystems
Tessil is a personal blog at tessil.github.io about programming and systems. Since 2017, 8 different Hacker News readers have linked it 9 times, most often “Benchmark of major hash maps implementations”. The last post was in 2017, but the archive is still widely read.
Too few sampled comments (5) to say how often HN recommends it.?
Why HN links it
HN readers link Tessil's 2016 hash map benchmark post (hopscotch-map) as a reference when comparing C++ hash and ordered map performance. Commenters cite it for its numbers, without judging the writing.
Summarised by Claude from a sample of 5 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- Benchmark of major hash maps implementations6 people · 7 mentions · first linked 2017-08-21
What Hacker News says?
sooo.... what's the difference with your run-of-the-mill hash map / ordered map / whatever ? How does it compare to other maps such as these ones ? https://tessil.github.io/2016/08/29/benchmark-hopscotch-map....
Comment linking /2016/08/29/benchmark-hopscotch-map.html
on “Faster – Fast key-value store from Microsoft Research”Submitted Benchmark of major hash maps implementations as a story · 2 points
HN discussionI don't understand the problems. Want a better hash map ? Use whatever makes your day here : https://tessil.github.io/2016/08/29/benchmark-hopscotch-map.... ; they are all header-only just like the standard library's. Want a vector that does not initialize its elements ? Ask and thou shalt receive: https://github.com/aguinet/pector The whole point of C++ is to make it easy for people to write generic libraries beyon…
Comment linking /2016/08/29/benchmark-hopscotch-map.html
on “Is C++ fast?”Submitted Benchmark of major hash maps implementations as a story · 2 points
HN discussionThat makes sense, thanks! For microbenchmarks, I try to remove any overheads during runtime by precomputing. That means fully populating the map (only hits) and generating the distribution of keys. Then the threads start at a random index and increment for the key on each operation. This is a good example in Java, https://github.com/ben-manes/caffeine/blob/master/caffeine/s... I would expect ~15-30ns per integer key…
Comment linking /2016/08/29/benchmark-hopscotch-map.html
on “Rust at speed – building a fast concurrent database [video]”Submitted HAT-trie, a cache-conscious trie as a story · 2 points
HN discussionIt's not very slow. It's faster than `std::map` and we lived with that for years! It's just not as fast as some other implementations: https://tessil.github.io/2016/08/29/benchmark-hopscotch-map....
Comment linking /2016/08/29/benchmark-hopscotch-map.html
on “Comparison of C++17, Go, and Java for a next-generation sequencing tool”Submitted Hopscotch Hashing as a story · 1 points
HN discussionhttps://tessil.github.io/2016/08/29/benchmark-hopscotch-map.... Has some older benchmarks, including those two.
Comment linking /2016/08/29/benchmark-hopscotch-map.html
on “A C++ implementation of a fast hash map and hash set using hopscotch hashing”