StronglyUniversalStringHashing

Really good and well written article. We did the similar trick with SSE. It's a good approach if a kind of non-portable hash can be used. In system I'm working on we introduced two APIs: hash() which maps to "fastest hash we have on this machine" and portable_hash() that's a predefined hash algorithm. BTW, SSE4 introduced carry-less multiplication which can also be used to boost hashing[1]. However, it's good for large data; in our tests it outperformed other algorithms for inputs larger than 1kB. [1] https://github.com/lemire/StronglyUniversalStringHashing

Where can I read what they mean by "Strong" hashing for HighwayHash? Also, is there a peer-reviewed paper on SipTreeHash? I am not a cryptographer, so I wouldn't trust my own instincts in terms of evaluating its strength compared to SipHash. For non-cryptographic applications, CLHASH produces 64 bits of result and is 2.004/2^64 almost delta universal, at a cost of 0.16 cycles per byte on Haswell and 0.10 cycles per byte on Skylake. https://github.com/lemire/StronglyUniversalStringHashing http://arxiv.org/abs/1503.03465

> I've literally never heard of Vhash, and I see no references to it outside of the 2007 paper. It is intimately tied up with VMAC. The title of the 2006 paper that introduced them is "Message authentication on 64-bit architectures". Google scholar says it has been cited 32 times, while "SipHash: a fast short-input PRF" has been cited 43 times. > As far as I know, the state of the art for "fast and secure" is still SipHash 2-4 or 2-3. SipHash seems to me like a good choice. SipHash has some security properties that Vhash does not, although Vhash-based-VMAC has similar security properties to SipHash, I believe. However, for hash flooding attacks, if the hash values themselves are not directly exposed to the attackers and there is enough noise in the response latency that timing attacks are difficult, I do not know what benefits SipHash has