Efficient semi-external depth-first search

• A novel and efficient semi-external DFS algorithm EP-DFS is presented. requires simpler CPU calculation less memory space. index devised to reduce the disk random accesses. Extensive experiments are conducted on both real synthetic datasets . As graphs grow in size, many real-world difficult load into primary of a computer. Thus, computing depth-first search (DFS) results (i.e., order or DFS-Tree) model important investigate. Semi-external algorithms assume that can at least hold spanning tree T graph G gradually restructure DFS-Tree, which nontrivial. In this paper, we present comprehensive study for problem. Based theoretical analysis problem, introduce new called with lightweight N + - Unlike traditional algorithms, focus addressing such complex problem efficiently fewer I/Os, calculations (implementation-friendly), I/O accesses (key-to-efficiency). experimental evaluations performed graphs, confirm proposed markedly outperforms existing algorithms.