HN Explorer

winwang

↗ HN profile [ 66.3 ms ] full profile
Karma
0
Created
()
Submissions
81
  1. Non-Messing-Up++: Diagonal Sorting and Young Tableaux (winwang.blog)
    14 points 1
  2. Google Cloud Console (GCP) Has Dark Mode (docs.cloud.google.com)
    1 points 1
  3. Ask HN: Which function definition keyword do you prefer, def or fn?

    I'm curious about what people think of `def` vs `fn`, two short keywords for starting function definitions. ...Or maybe we should subscribe to the church of `df`, haha.

    1 points 5
  4. Faster sorting with SIMD CUDA intrinsics (2024) (winwang.blog)

    Code at https://github.com/wiwa/blog-code/

    92 points 11
  5. Ask HN: People using BigQuery for data ingestion, biggest pain points?

    Specifically for batch ETL/ELT workloads, assuming 10s/100s of TB.

    2 points 0
  6. Can GenAI Improve Developer Productivity? (resources.uplevelteam.com)
    1 points 1
  7. 2 points 1
  8. Leveraging Spark 3 and Nvidia's GPUs to Reduce Cloud Cost by Up to 70% (medium.com)
    3 points 1
  9. RegDem: Increasing GPU Performance via Shared Memory Register Spilling (arxiv.org)
    2 points 1
  10. Show HN: Beating Microsoft's Garnet KV Store in less than 300 lines of code (github.com)

    I saw Microsoft's Garnet getting to the limits of (CPU) concurrent hashmap performance and thought it would be interesting to see if I could beat it with a GPU. Specifically, the GET throughput at 256M keys. Of course,…

    3 points 2
  11. Show HN: Faster sorting with register shuffling in CUDA (github.com)

    Hi! I recently posted a blog post around this technique (i.e. using SIMD-like warp intrinsics), but I wanted to show the runnable code. It should be fairly portable. tl;dr: Warp-wide (32-way) bitonic mergesort is…

    3 points 0
  12. Sorting with CUDA (SIMD) Shuffling (winwang.blog)
    4 points 1
  13. Can we 10x Rust hashmap throughput? (wiwa.substack.com)
    11 points 0