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

Groups > comp.programming > #4139 > unrolled thread

Here is the benchmarks...

Back to article view | Back to comp.programming

Contents

  Here is the benchmarks... aminer <aminer@toto.net> - 2014-03-18 01:14 -0700
    Re: Here is the benchmarks... aminer <aminer@toto.net> - 2014-03-18 01:18 -0700
    Re: Here is the benchmarks... aminer <aminer@toto.net> - 2014-03-18 01:24 -0700
    Re: Here is the benchmarks... aminer <aminer@toto.net> - 2014-03-18 01:45 -0700

#4139 — Here is the benchmarks...

Hello,


I have benchmarked cache-lines transfers etc. and i have
done all the calculations and here is the results:


On a computer with 32 cores LW_RWLockX scales up to 20X


On a computer with 32 cores LW_RWLock scales up to 20X


And on a computer with 32 cores my RWLock and my RWLockX scales up to 5X.


So i advise you to use LW_RWLockX and LW_RWLock that scales better.



You can download all the variant of my scalable RWLock 3.05 from:


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



Thank you,
Amine Moulay Ramdane.

[toc] | [next] | [standalone]

#4140

Hello,


Those benchmarks was calculated for an x86 computer with 32 cores.




Thank you,
Amine Moulay Ramdane.

[toc] | [prev] | [next] | [standalone]

#4141

Hello,


Those benchmarks was calculated for a scenario of 0.1% of writes.




Thank you,
Amine Moulay Ramdane.

[toc] | [prev] | [next] | [standalone]

#4142

Hello,

On my benchmarks and calculations the execution time under the read 
section and write section are the same, i have done it like
that to show you to how much can they scale.



Thank you,
Amine Moulay Ramdane.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.programming

csiph-web