Review

[tkaitchuck]

Review of aHash is needed both in implementation and conceptually to ensure that it actually satisfies the properties it is attempting to guarantee.

[tkaitchuck]

This may serve as a starting point https://github.com/tkaitchuck/aHash/wiki/How-aHash-is-resists-DOS-attacks

[ciphergoth]

In order to do this, we could do with a short paper precisely specifying the algorithms you want reviewed! You say something about "not cryptographically secure", but what you want is a cryptographic property - that an attacker who doesn't know the secret key will find it difficult to create a lot of collisions in the hash table. It would be interesting to try to specify this more precisely.