Top all-time · #22384
Programming and mathematics
Personal blog · cachemiss.xyz/blogprogrammingmath
Programming and mathematics is a personal blog at cachemiss.xyz about programming and math. Since 2020, 3 different Hacker News readers have linked it 3 times, most often “Parallel Reduce and Scan on the GPU”. It is still writing: the last post was on 2026-03-31, and you can follow it by feed.
Too few sampled comments (2) to say how often HN recommends it.?
Why HN links it
Posts on GPU compute, covering Vulkan prefix sum and parallel reduce and scan implementations. Commenters link them as background on prefix sums and compare their approaches; no comment evaluates the writing.
Summarised by Claude from a sample of 2 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- Parallel Reduce and Scan on the GPU3 people · 3 mentions · first linked 2020-04-17
What Hacker News says?
That sounds like a great project, one I heartily recommend to people wanting to get their feet wet. First, the basic principles of how to write an efficient prefix sum are the same as before. Definitely read [1]. Also see [2] for an exploration into how much subgroups help over threadgroup shared memory. In fact, [2] has much of the answer you seek, as it's already written in Vulkan and GLSL (compiling to SPIR-V), t…
Comment linking /blog/parallel-reduce-and-scan-on-the-GPU
on “A Taste of GPU Compute [video]”Submitted Parallel Reduce and Scan on the GPU as a story · 16 points
HN discussionI previously read this: https://cachemiss.xyz/blog/parallel-reduce-and-scan-on-the-G... whose implementation seems a lot simpler than what is presented here. What is the main difference here? I've only dabbled a bit in Vulkan. (I wrote a pathtracer that I wanted to speed up using compaction; for which I needed a prefix sum)
Comment linking /blog/parallel-reduce-and-scan-on-the-GPU
on “Prefix Sum on Vulkan”