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

Groups > comp.programming.threads > #2278 > unrolled thread

More information...

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

Contents

  More information... aminer <aminer@toto.net> - 2014-04-27 15:24 -0700

#2278 — More information...

Hello,


I have said yesterday that my algorithm of a concurrent FIFO queue that 
uses a lockfree mechanism on the pop() scored less than the Chriss 
Thomasson algorithm, here it is:

http://pages.videotron.com/aminer/CQueue1.htm


And that's not true, i have not used yesterday a sleep(0), so the 
lockfree CAS loop was genrating much more "contention", i have put a 
sleep(0) after that the CAS fails and it has given 5.2 millions of pop 
transactions per second on under contention with 4 threads on my 2.4 GHz 
Quadcore, the Chriss Thomasson algorithm has given 4.8 millions of 
transactions per second on the pop side, on the push side they have the 
same throughput.

But my algorithm with a lockfree CAS on the pop is not starvation-free , 
Chriss Thomasson algorithm is starvation-free.



Thank you,
Amine Moulay Ramdane.

[toc] | [standalone]

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

csiph-web