<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
  <title>Lilje Due's Blog</title>
  <link href="https://liljedue.dk/content/feed.xml" rel="self"/>
  <link href="https://liljedue.dk/"/>
  <id>https://liljedue.dk/</id>
  <updated>2026-11-02T00:00:00Z</updated>
    <entry>
    <title>How To Parallelize Your Tree Operations</title>
    <link href="https://liljedue.dk/blog/parallel-trees"/>
    <id>https://liljedue.dk/blog/parallel-trees</id>
    <updated>2026-11-02T00:00:00Z</updated>
    <author><name>Lilje Henrich Due</name></author>
    <summary>Computations on trees are often seen as inherently sequential. In this post, I will show how to compute things like node depths and subtree sizes in parallel.</summary>
  </entry>  <entry>
    <title>What Is The Best Parallel Partition?</title>
    <link href="https://liljedue.dk/blog/parallel-partition"/>
    <id>https://liljedue.dk/blog/parallel-partition</id>
    <updated>2026-08-31T00:00:00Z</updated>
    <author><name>Lilje Henrich Due</name></author>
    <summary>How to implement partition efficiently on a GPU in Futhark, and why there is no single best approach.</summary>
  </entry>  <entry>
    <title>Turning an Inclusive Scan into an Exclusive Scan</title>
    <link href="https://liljedue.dk/blog/incscan-to-exscan"/>
    <id>https://liljedue.dk/blog/incscan-to-exscan</id>
    <updated>2026-06-15T00:00:00Z</updated>
    <author><name>Lilje Henrich Due</name></author>
    <summary>Using monoid composition it is possible to turn an inclusive scan into an exclusive scan.</summary>
  </entry>  <entry>
    <title>Monoid Composition &amp; Maximum Subarray Sum Problem</title>
    <link href="https://liljedue.dk/blog/monoid-composition"/>
    <id>https://liljedue.dk/blog/monoid-composition</id>
    <updated>2026-06-14T00:00:00Z</updated>
    <author><name>Lilje Henrich Due</name></author>
    <summary>Some monoids can be composed into a single monoid, allowing a scan feeding into a reduce to be expressed as a single reduce. This turns out to be a useful combinator, with applications in parentheses matching, polynomial evaluation, and the maximum subarray sum problem.</summary>
  </entry>  <entry>
    <title>Parallel Stacks &amp; Parentheses Matching</title>
    <link href="https://liljedue.dk/blog/parallel-parentheses-matching"/>
    <id>https://liljedue.dk/blog/parallel-parentheses-matching</id>
    <updated>2026-06-02T00:00:00Z</updated>
    <author><name>Lilje Henrich Due</name></author>
    <summary>The parallel parentheses matching problem can be parallelized in different manners depending on the use case and is a good example of how stacks map to parallel hardware.</summary>
  </entry>
</feed>