Re: HashMap

[Greg Brown]

I have made a few optimizations and fixes. The test no longer fails  
for 100,000 puts() - it now takes ~0.65s. However, java.util.HashMap  
does it in ~0.2s, so there is still room for improvement. I think the  
current bottleneck may be in the put() method itself, which is called  
to perform each individual put() as well as to re-put() the current  
contents of the map when we re-hash.

It is probably related to how we are hashing internally -  
java.util.HashMap's hashing scheme is undoubtedly more highly  
optimized. I'll keep digging.