Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.programming.threads > #2304 > unrolled thread

Please read again...

Back to article view | Back to comp.programming.threads

Contents

  Please read again... aminer <aminer@toto.net> - 2014-05-05 21:12 -0700
    Re: Please read again... Kaz Kylheku <kaz@kylheku.com> - 2014-05-07 01:47 +0000

#2304 — Please read again...

Hello,


Please read again about my new algorithm of a very fast concurrent FIFO 
queue, i have just correted some typos, here it is:

http://pages.videotron.com/aminer/CQueue3.htm


My new algorithm has more parallelism than my previous algorithms, and 
it minimizes efficiently the cache-coherence traffic, and it scales 
better even if the number of threads are greater than the number of 
cores and my new algorithm has scored 20% more throughput on the pop() 
side than the following Chriss Thomasson algorithm that uses the bakery 
algorithm and the same throughput on the push() side than the following 
Chriss Thomasson algorithm.


Here is the Chriss Thomasson algorithm:

https://groups.google.com/forum/#!topic/lock-free/acjQ3-89abE/discussion


Thank you,
Amine Moulay Ramdane.

[toc] | [next] | [standalone]

#2307

On 2014-05-06, aminer <aminer@toto.net> wrote:
>
> Hello,
>
> Please read again about my new algorithm of a very fast concurrent FIFO 
> queue, i have just correted some typos, here it is:

What real-world application benefits from this?

How many percent faster is the overall performance of that application,
compared to using a slow reference FIFO implemenation?

Can you make some useful program run 10 to 15 percent faster with this?

[toc] | [prev] | [standalone]

Back to top | Article view | comp.programming.threads

csiph-web