A fast, zero-dependency implementation of bitwise Hamming Distance using a method amenable to auto-vectorization.

This started out as a benchmark of various bitwise Hamming distance implementations in Rust. However, after finding that a simple implementation that is amenable to auto-vectorization was comparable, if not faster, than other implementations, I decided to publish it as a crate.

Note: This is for comparing bit-vectors, not for comparing strings.

use hamming_bitwise_fast::hamming_bitwise_fast;

assert_eq!(hamming_bitwise_fast(&[0xFF; 1024], &[0xFF; 1024]), 0);
assert_eq!(hamming_bitwise_fast(&[0xFF; 1024], &[0x00; 1024]), 1024);

This uses Criterion to benchmark various Hamming distance implementations:

• The auto-vectorized implementation in this crate
• A naive for-loop based implementation
• A naive iterator based implementation
• hamming
• hamming_rs
• simsimd

cargo bench

Then open the target/criterion/report/index.html file in your browser to view the results.

These were the results running on 3 different types of machines:

This project is licensed under either of the following licenses, at your option:

• Apache License, Version 2.0
• MIT License

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in this project by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.