Prev: Programming v Mathematics
Next: About Threadpool...
From: aminer on 1 Mar 2010 11:25

Hello,


Description:


Parallel algorithm to handle queue FIFO that reduce contention
and enhance the parallel throuput of the lockfree_mpmc pop() method,
it can go now up to 7.037 millions of transactions per second
on an Intel Core 2 Quad Q6600


Language: FPC Pascal v2.2.0+ / Delphi 5+: http://www.freepascal.org/


Operating Systems: Win , Linux and Mac (x86).


For ParallelQueue benchmarks on an Intel Core 2 Quad Q6600 look at:


http://pages.videotron.com/aminer/parallelqueue/parallelqueue.htm


I have tested 3 lockfree queue algorithms:


flqueue at: http://www.emadar.com/fpc/lockfree.htm


RingBuffer at: http://www.odsrv.com/RingBuffer/RingBuffer.htm


ParallelQueue at: http://pages.videotron.com/aminer/


Sincerely,
Amine Moulay Ramdane.

From: aminer on 1 Mar 2010 22:05

Hello,

And don't forget to compile with -Sd (delphi mode) when using FPC.



Regards,
Amine.






On Mar 1, 11:25 am, aminer <ami...(a)videotron.ca> wrote:
> Hello,
>
> Description:
>
> Parallel algorithm to handle queue FIFO that reduce contention
> and enhance the parallel throuput of the lockfree_mpmc pop() method,
> it can go now up to 7.037 millions of transactions per second
> on  an Intel Core 2 Quad Q6600
>
> Language: FPC Pascal v2.2.0+ / Delphi 5+:http://www.freepascal.org/
>
> Operating Systems: Win , Linux and Mac (x86).
>
> For ParallelQueue benchmarks on an Intel Core 2 Quad Q6600 look at:
>
> http://pages.videotron.com/aminer/parallelqueue/parallelqueue.htm
>
> I have tested 3 lockfree queue algorithms:
>
> flqueue at:http://www.emadar.com/fpc/lockfree.htm
>
> RingBuffer at:http://www.odsrv.com/RingBuffer/RingBuffer.htm
>
> ParallelQueue at:http://pages.videotron.com/aminer/
>
> Sincerely,
> Amine Moulay Ramdane.

 | 
Pages: 1
Prev: Programming v Mathematics
Next: About Threadpool...