Adaptive Hashing

https://quotenil.com/adaptive-hashing.html

by varjag • 1 year ago

56 8 1 year ago

tptacek2• 1 year ago

Fun fact! A previous use of the term "adaptive hash" was as a descriptor for things like Bcrypt, which have the exact opposite goal (to consistently be as slow as possible regardless of advances in hardware).

90s_dev• 1 year ago

"tptacek's law: given enough time, diametrically opposed algorithms will have the same name"

tptacek2• 1 year ago

Yes! See: quicksort.

Nevermark• 1 year ago

So we just flip a sign in Bcrypt! Or flip a decision Boolean? Or a sort order?

Regardless it will be easy. Apply the inverse operation of the “introduce more slowness” operation.

Unfortunately, I have seen some software in my day, and I don’t think it will work. Just really perverse stuff.

vlovich123• 1 year ago

For string keys, hash only the first and last 2 characters. For list keys, only hash the first 4 elements.

I would think it make some sense to change these constants as the collision chain grows, thereby improving the hash quality of keys and avoiding collisions as the table gets larger and larger

BoingBoomTschak• 1 year ago

If too many collisions, double the limit.

heisig• 1 year ago

Let me comment as an SBCL user: This is outstanding work, and I can now remove a lot of performance hacks from my code because the default hash tables became equally fast!

Also, this technique eliminates a number of worst-case scenarios and inefficiencies, which is a boon for any hash table user.

Adaptive Hashing

Related Stories