New method cache, 30% faster macrobenchmarks and ineffeciencies.

Stephan Rudlof sr at evolgo.de
Mon Dec 10 02:19:10 UTC 2001


Scott A Crosby wrote:
> 
> On Sun, 9 Dec 2001, Stephan Rudlof wrote:
> 
> 
> > But for ordinary - not VM related - hashed collections an alternative hash
> > scheme with VM support (named #longHash?) could make sense. And the use of
> 
> Its probably not relevant. My hashed collection stuff avoids a nasty big
> performance degradation for when you get to >4000 elements (where
> performace will fall by a factor of 1000 as the algorithm devolves into a
> linear search).. If someone needs to store >100k objects in a hash table,
> they'll probably be far better served by writing a custom 'hash' function
> and store a 30-bit hash value in a slot.

I think I've taken the point: spreading a 12 bit hash to *much* larger
values doesn't help for very large collections: it leads to big leaks in the
collections just reachable by linear search...

Thanks,

Stephan


PS: now I've learned a lot about hash values...

> 
> Scott

-- 
Stephan Rudlof (sr at evolgo.de)
   "Genius doesn't work on an assembly line basis.
    You can't simply say, 'Today I will be brilliant.'"
    -- Kirk, "The Ultimate Computer", stardate 4731.3




More information about the Squeak-dev mailing list