chenxuhao / gpuscc_code

Parallel identification of strongly connected components on GPU

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

usage: ./scc [-t] [-a t/d/l/h] -f graph
	-t tarjan's algorithm, optional.
	-a q for topology-driven, d for data-driven, l for load-balance, h for hybrid. topology-driven in default if not specific.
	-f input directed graph

apps: scc scc_w scc_2_w
scc:	parallel FB implementation on CUDA
scc_w:	scc with finding wcc once
scc_2_w	scc_w with trim_2 once

About

Parallel identification of strongly connected components on GPU

License:MIT License


Languages

Language:Cuda 80.6%Language:C++ 10.4%Language:C 7.6%Language:Makefile 1.2%Language:Shell 0.2%