Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Implement parallel is_heap and is_heap_until. #2654
I writed benchmark codes. And then I found that my implementation's performance is better than std::* and has scalability as the number of cores increase.
Benchmark for is_heap_until (Fix the number of cores and adjust N)
Benchmark for is_heap_until (Fix N and adjust M(the number of cores used))
How about you add it here: https://github.com/STEllAR-GROUP/hpx/tree/master/tests/performance/local?
@hkaiser I'll add many performance tests in summer. I think creating new folder and put them into it is better. So which path is good do you think?
May 31, 2017
1 check passed
In above code, accessing parent node is bad to cache. But I have no solutions to resolve that problem with keeping current algorithm for
If anyone want to get better performance in
This is a note for someone in the future.