[concurrency-interest] Is lock-free list useful?

James Gan ganzhi at gmail.com
Wed Mar 3 09:28:46 EST 2010


Roman,

I read following article when I was creating lock-free set and list:

http://www.research.ibm.com/people/m/michael/spaa-2002.pdf
High Performance Dynamic Lock-Free Hash Tables and List-Based Sets
Maged M. Michael

If you are interested in my implementation, here is the link:
http://amino-cbbs.svn.sourceforge.net/viewvc/amino-cbbs/trunk/amino/java/src/main/java/org/amino/ds/lockfree/LockFreeList.java?revision=555&view=markup

On Wed, Mar 3, 2010 at 8:34 PM, Roman Elizarov <elizarov at devexperts.com> wrote:
> Hello James!
>
> Right. Its underlying Michael-Scott algorithm does not support this
> operation. Then, what simple algorithm for lock-free list you are
> referring to? Can you give a reference to a paper or article about it,
> please (the name of the paper and its authors will be enough).
>
> Sincerely,
> Roman Elizarov
>
> On Wednesday, March 3, 2010 3:07:40 PM you wrote:
>
> JG> emm, ConcurrentLinkedQueue didn't support "insertion in the middle"
>
> JG> On Wed, Mar 3, 2010 at 5:18 PM, Roman Elizarov <elizarov at devexperts.com> wrote:
>>> Hello James!
>>>
>>> JG> I'm feeling interesting that there is no lock-free list in the j.u.c
>>> JG> package. Is there a reason behind this? The algorithm of lock-free
>>> JG> list is pretty simple. And list is a very useful data structure.
>>>
>>> java.util.concurrent.ConcurrentLinkedQueue
>>>
>>> Sincerely,
>>> Roman Elizarov
>>>
>>>
>
>
>
>
>
>
>



-- 
Best Regards
James Gan
Current Project: Concurrent Building Block at http://amino-cbbs.sourceforge.net/
Blog: http://ganzhi.blogspot.com


More information about the Concurrency-interest mailing list