Performance analysis and optimization on HPX Rotate algorithm

HPX is a C++ Standard Library for Parallelism and Concurrency, which implements parallel versions of all C++ Standard algorithms. This poster presents the optimization of the HPX parallel rotate algorithm. The original implementation failed to effectively distribute work among CPU cores and caused un-needed overhead due to redundant usage of a hpx::future. The optimization consists of introducing core partitioning functionality and removing unnecessary overhead. The results show significant performance gain, averaging a 2x improvement for input sizes smaller than 2^22 compared to the original implementation.

Related Organizations

Louisiana State University
United States
Aristotle University of Thessaloniki
Greece

Keywords

HPX parallel algorithm, rotate algorithm, performance optimization, core partition functionality, HPX parallel algorithm, performance optimization, core partition functionality

Impact byBIP!

	selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	0
	popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.	Average
	influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).	Average
	impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.	Average