[Collections] A better TreeList?

2013-10-09 Thread Rodion Efremov

Hello, y'all!

My buggy experiments show that it makes sense to produce a 
TreeList-like structure, that doesn't use an entire AVLNode for storing 
each god damned element, but rather stores, say, 128 elements in a node. 
In such a way, you can talk about a structure that takes the best from 
Array-/Linked-/Tree-list.


Your opinion?

--
TIA, Rodion



-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org



Re: [Collections] A better TreeList?

2013-10-09 Thread Bernd Eckenfels

Am 09.10.2013, 11:24 Uhr, schrieb Rodion Efremov :
My buggy experiments show that it makes sense to produce a TreeList-like  
structure, that doesn't use an entire AVLNode for storing each god  
damned element, but rather stores, say, 128 elements in a node. In such  
a way, you can talk about a structure that takes the best from  
Array-/Linked-/Tree-list.


Just a related note, JEP-180 hit OpenJDK, they introduced a balanced tree  
implementation for hash collisions (strange enough the implementation does  
not used a array based implementation)


Bernd

-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org



Re: [Collections] A better TreeList?

2013-10-09 Thread James Carman
This is basically a b-tree, the "TreeList-like structure."

On Wed, Oct 9, 2013 at 2:15 PM, Bernd Eckenfels  wrote:
> Am 09.10.2013, 11:24 Uhr, schrieb Rodion Efremov :
>>
>> My buggy experiments show that it makes sense to produce a TreeList-like
>> structure, that doesn't use an entire AVLNode for storing each god damned
>> element, but rather stores, say, 128 elements in a node. In such a way, you
>> can talk about a structure that takes the best from
>> Array-/Linked-/Tree-list.
>
>
> Just a related note, JEP-180 hit OpenJDK, they introduced a balanced tree
> implementation for hash collisions (strange enough the implementation does
> not used a array based implementation)
>
> Bernd
>
> -
> To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> For additional commands, e-mail: dev-h...@commons.apache.org
>

-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org