Data Processing Benchmark (60k Posts)

Top 5 related posts generation based on shared tags.

Methodology

Timings taken on an AWS c7a.xlarge (4vCPU, 8GB RAM). The benchmark processes a JSON dataset of 60,000 posts and calculates relationships.

• Julia HO: Highly optimized version using specialized data structures.
• D HO: Highly optimized version.
• JS runtimes: Comparison between Bun, Node, and Deno.
• Legacy: Python and Ruby showing interpreted overhead.

Raw Output (60k Dataset)

Language          Time
Julia HO          99.33 ms
D HO              122.06 ms
Rust              1.30 s
Zig               1.99 s
Go                3.48 s
Swift             4.19 s
JS (Bun)          8.49 s
JS (Node)         11.03 s
Python            215.18 s
Ruby              254.93 s