diff --git a/year2024/src/day09.rs b/year2024/src/day09.rs
new file mode 100644
index 0000000..7196964
--- /dev/null
+++ b/year2024/src/day09.rs
@@ -0,0 +1,161 @@
+use std::{cmp::Reverse, collections::BinaryHeap};
+
+use common::day::Day;
+
+pub struct Day09 {}
+
+impl Day for Day09 {
+    fn star1(&self, input: &str) -> String {
+        let mut blocks = parse_input_star1(input);
+        compact(&mut blocks);
+        format!("{}", checksum_star1(&blocks))
+    }
+
+    fn star2(&self, input: &str) -> String {
+        let (mut file_offsets, mut free_offsets) = parse_input_star2(input);
+        defrag(&mut file_offsets, &mut free_offsets);
+        format!("{}", checksum_star2(&file_offsets))
+    }
+}
+
+fn parse_input_star1(input: &str) -> Vec<Option<usize>> {
+    let mut blocks = vec![];
+
+    for (i, c) in input.trim().chars().enumerate() {
+        let length = c.to_digit(10).unwrap();
+        let entry = if i % 2 == 0 { Some(i / 2) } else { None };
+
+        for _ in 0..length {
+            blocks.push(entry);
+        }
+    }
+
+    blocks
+}
+
+fn checksum_star1(blocks: &[Option<usize>]) -> usize {
+    blocks
+        .iter()
+        .enumerate()
+        .map(|(i, c)| i * c.unwrap_or(0))
+        .sum()
+}
+
+fn compact(blocks: &mut [Option<usize>]) {
+    let mut free_pos = blocks.iter().position(|e| e.is_none()).unwrap();
+    let mut file_pos = blocks.len() - 1 - blocks.iter().rev().position(|e| e.is_some()).unwrap();
+    assert_eq!(file_pos, blocks.len() - 1);
+
+    loop {
+        blocks.swap(free_pos, file_pos);
+
+        // get next file block
+        loop {
+            file_pos -= 1;
+            if blocks[file_pos].is_some() {
+                break;
+            }
+        }
+
+        // get next free block
+        loop {
+            free_pos += 1;
+            if blocks[free_pos].is_none() {
+                break;
+            }
+        }
+
+        if free_pos > file_pos {
+            return;
+        }
+    }
+}
+
+struct File {
+    offset: usize,
+    len: usize,
+}
+
+fn parse_input_star2(input: &str) -> (Vec<File>, [BinaryHeap<Reverse<usize>>; 9]) {
+    let mut free_offsets = [const { BinaryHeap::new() }; 9];
+    let mut file_offsets = vec![];
+
+    let mut offset = 0;
+
+    for (i, c) in input.trim().chars().enumerate() {
+        let len = c.to_digit(10).unwrap() as usize;
+        if i % 2 == 0 {
+            file_offsets.push(File { offset, len });
+        } else if len > 0 {
+            free_offsets[len - 1].push(Reverse(offset));
+        }
+        offset += len;
+    }
+
+    (file_offsets, free_offsets)
+}
+
+fn defrag(file_offsets: &mut [File], free_offsets: &mut [BinaryHeap<Reverse<usize>>; 9]) {
+    for file in file_offsets.iter_mut().rev() {
+        let mut chosen = None;
+
+        // find the free space interval which will fit the file, which has the lowest offset
+        // search only the buckets which are at least this size
+        for l in file.len..=9 {
+            if let Some(Reverse(offset)) = free_offsets[l - 1].peek() {
+                if *offset < file.offset {
+                    if let Some((_, chosen_offset)) = chosen {
+                        if offset < chosen_offset {
+                            chosen = Some((l, offset));
+                        }
+                    } else {
+                        chosen = Some((l, offset));
+                    }
+                }
+            }
+        }
+
+        // we have found a fitting free space
+        if let Some((l, new_offset)) = chosen {
+            // reset the file offset to the free space
+            file.offset = *new_offset;
+
+            // remove the now occupied free space
+            free_offsets[l - 1].pop();
+            // if there is any remaining free space, add it as a new free space to the correct bucket
+            if l > file.len {
+                let diff = l - file.len;
+                free_offsets[diff - 1].push(Reverse(file.offset + file.len));
+            }
+        }
+    }
+}
+
+fn checksum_star2(file_offsets: &[File]) -> usize {
+    let mut sum = 0;
+    for (id, file) in file_offsets.iter().enumerate() {
+        for i in file.offset..(file.offset + file.len) {
+            sum += i * id;
+        }
+    }
+    sum
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    const INPUT: &str = r#"2333133121414131402"#;
+
+    #[test]
+    fn star1() {
+        let d = Day09 {};
+        assert_eq!(d.star1(INPUT), "1928");
+    }
+
+    #[test]
+    fn star2() {
+        let d = Day09 {};
+        assert_eq!(d.star2(INPUT), "2858");
+    }
+}
diff --git a/year2024/src/lib.rs b/year2024/src/lib.rs
index b08c18f..55f09a0 100644
--- a/year2024/src/lib.rs
+++ b/year2024/src/lib.rs
@@ -9,6 +9,7 @@ mod day05;
 mod day06;
 mod day07;
 mod day08;
+mod day09;
 
 pub struct Year2024 {}
 
@@ -23,6 +24,7 @@ impl Year for Year2024 {
             6 => Some(Box::new(day06::Day06 {})),
             7 => Some(Box::new(day07::Day07 {})),
             8 => Some(Box::new(day08::Day08 {})),
+            9 => Some(Box::new(day09::Day09 {})),
             _ => None,
         }
     }