Same proof, less work.
This pow solver has been used by krautstrike instead of the slow python script. A pow-as-a-service deployment has been provided by tu-berlin and @domenukk.
Make sure you build a release binary with cargo build --release.
$ cargo test
$ cargo +nightly bench --features=nightly
running 5 tests
test tests::test_check_pow_invalid ... ignored
test tests::test_check_pow_valid ... ignored
test tests::test_pow_hash ... ignored
test tests::bench_pow_hash ... bench: 411 ns/iter (+/- 92)
test tests::bench_pow_valid ... bench: 1,459 ns/iter (+/- 223)
test result: ok. 0 passed; 0 failed; 3 ignored; 2 measured; 0 filtered out
$