On 216k bodies (N=50 in Sergei's script) and 3 threads, the total speedup is x1.75! The collider speedup is x3.5, but more time is spent in interactions (since verletDist is higher): x0.72 speed-down. Performance should be even better with more threads, since the collider/dispatcher are better balanced (50/50 with 3 threads). I'm currently testing 4 threads.
-- You received this bug notification because you are a member of Yade developers, which is the registrant for Yade. https://bugs.launchpad.net/bugs/729079 Title: Performance optimization of InsertionSortCollider Status in Yet Another Dynamic Engine: New Bug description: Sergei Dorofeenko (https://launchpad.net/~sergei.dorofeenko) found, that InsertionSortCollider is probably is a "bottle neck" in simulations with >10^5 number of particles even in many-threads mode. http://www.mail-archive.com/[email protected]/msg06573.html Сitation: "...I did a perfomance test for parallel mode and results in no good. Performance boost only about 40% from 1 thread to 4 thread for 200k particles... Cause is a non-parallelised InsertionSortCollider, who need about 80% time with 4 threads. Results attached." _______________________________________________ Mailing list: https://launchpad.net/~yade-dev Post to : [email protected] Unsubscribe : https://launchpad.net/~yade-dev More help : https://help.launchpad.net/ListHelp
