Faster HashMap implementation

Ismael Juma mlists at
Sat Jun 13 11:00:39 UTC 2009

Hi Doug,

Doug Lea <dl at ...> writes:
> While open-addressing is used in
> IdentityHashMap (and in a very specialized form in
> ThreadLocal), you cannot live with linear-probed
> versions otherwise: Many user-defined hashCodes
> (and some JDK-defined ones too!) are not very good.

Out of curiosity, do you know if any tests have been done with an open
addressing scheme similar to Python dictionaries[1] in Java? Near the top,
there's a comment explaining how it works and the motivation.



More information about the core-libs-dev mailing list