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

Groups > comp.programming.threads > #887 > unrolled thread

About my parallelsort library ...

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

Contents

  About my parallelsort library ... "aminer" <aminer@videotron.ca> - 2012-05-20 12:19 -0500
    Re: About my parallelsort library ... "aminer" <aminer@videotron.ca> - 2012-05-20 13:02 -0500

#887 — About my parallelsort library ...

Hello,

If you take a look at the graph of the benchmarks that i did on
my parallelsort library , you will notice that i am getting a superlinear
speed up , look at this:

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


So a question follows:

How am i getting this superlinear speed up on this example using
10 millions of strings?


My answer:

Cause when you are using mutliple threads and mutiple cores,
my parallelsort library is maximizing BETTER the locality , that means
the caches are used more efficiently when more threads and more cores
are used .



You can download my parallelsort library from:


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



Thank you.


Amine Moulay Ramdane

[toc] | [next] | [standalone]

#888

I wrote:

>Cause when you are using mutliple threads and mutiple cores, my 
>parallelsort library is maximizing BETTER the locality

I mean exploit better the cache locality.


thank you.



Amine Moulay Ramdane.



"aminer" <aminer@videotron.ca> wrote in message 
news:jpb5jg$nr2$1@dont-email.me...
>
> Hello,
>
> If you take a look at the graph of the benchmarks that i did on
> my parallelsort library , you will notice that i am getting a superlinear
> speed up , look at this:
>
> http://pages.videotron.com/aminer/parallelsort/parallelsort.htm
>
>
> So a question follows:
>
> How am i getting this superlinear speed up on this example using
> 10 millions of strings?
>
>
> My answer:
>
> Cause when you are using mutliple threads and mutiple cores,
> my parallelsort library is maximizing BETTER the locality , that means
> the caches are used more efficiently when more threads and more cores
> are used .
>
>
>
> You can download my parallelsort library from:
>
>
> http://pages.videotron.com/aminer/
>
>
>
> Thank you.
>
>
> Amine Moulay Ramdane
>
>
>
>
>
> 

[toc] | [prev] | [standalone]

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

csiph-web