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

Groups > comp.programming.threads > #2786 > unrolled thread

About my new parallel sort algorithm

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

Contents

  About my new parallel sort algorithm Ramine <ramine@1.1> - 2015-01-01 21:47 -0800

#2786 — About my new parallel sort algorithm

Hello,


As i have told you , my new parallel Sort algorithm has become
more cache-aware, and since it has become more cache-aware it
have induced a super linear speedup and sper linear scalability when 
using more cores and more L2 caches, i have done some benchmarks on my 
Quadcore that uses two L2 caches and it has given a super linear speedup 
of 5X scalability on my Quadcore when sorting strings even though i am 
using only 4 cores, that's easy to understand cause when you use only 
one thread it will use only one L2 cache, but when you use more threads 
on multiple cores and with multiple L2 caches it will use more L2 caches 
and it will parallelize the access to those multiple L2 caches , this is 
why my new parallel algorithm has given a super linear speedup when 
sorting strings.

So be happy with my new parallel Sort library:

You can download my new Parallel Sort library version 3.3 from:

https://sites.google.com/site/aminer68/parallel-sort-library



Thank you,
Amine Moulay Ramdane.

[toc] | [standalone]

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

csiph-web