<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Parallel-Programming on Yi-Wei Lien</title>
    <link>https://lien0214.github.io/en/tags/parallel-programming/</link>
    <description>Recent content in Parallel-Programming on Yi-Wei Lien</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Sun, 14 Jun 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://lien0214.github.io/en/tags/parallel-programming/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>We Tried to Make Seam Carving Fast on a GPU. Then We Proved Why It Can&#39;t Be.</title>
      <link>https://lien0214.github.io/en/posts/seam-carving-gpu/</link>
      <pubDate>Sun, 14 Jun 2026 00:00:00 +0000</pubDate>
      <guid>https://lien0214.github.io/en/posts/seam-carving-gpu/</guid>
      <description>A profiling-driven study of exact seam carving on an NVIDIA V100 — optimizing Naive → Fused → Prefetch → BytePtr → Tiled, reaching 5.73 ms/seam at 8K, and then proving the ceiling is structural.</description>
    </item>
  </channel>
</rss>
