mkscrg / hashring Goto Github PK

In the comments in the code you claim O(1) lookup performance. I don't believe this is correct or even possible.

In lookup, you're using IntMap.splitLookup. This is O(log n) which is what I'd expect given that it's a tree structure. But that would correspond to O(NR) if N is the number of nodes added.

Maybe I'm missing something, but I can't see how your lookups (or indeed any space-efficient implementation of CH) can be O(1).

Hey Mike,

Just noticed your library while researching for a consistent hashing solution. Looks real nice!

I do have one question though - isn't it a problem that the hashable package does not promise to produce consistent values across program runs? Wouldn't this mean that 3 separate nodes looking at the same list of servers may find their messages in different nodes?

Cheers,
Oz