English

$2$-Fault-Tolerant Strong Connectivity Oracles

Data Structures and Algorithms 2023-12-05 v2

Abstract

We study the problem of efficiently answering strong connectivity queries under two vertex failures. Given a directed graph GG with nn vertices, we provide a data structure with O(nh)O(nh) space and O(h)O(h) query time, where hh is the height of a decomposition tree of GG into strongly connected subgraphs. This immediately implies data structures with O(nlogn)O(n \log{n}) space and O(logn)O(\log{n}) query time for graphs of constant treewidth, and O(n3/2)O(n^{3/2}) space and O(n)O(\sqrt{n}) query time for planar graphs. For general directed graphs, we give a refined version of our data structure that achieves O(nm)O(n\sqrt{m}) space and O(m)O(\sqrt{m}) query time, where mm is the number of edges of the graph. We also provide some simple BFS-based heuristics that seem to work remarkably well in practice. In the experimental part, we first evaluate various methods to construct a decomposition tree with small height hh in practice. Then we provide efficient implementations of our data structures, and evaluate their empirical performance by conducting an extensive experimental study on graphs taken from real-world applications.

Keywords

Cite

@article{arxiv.2311.00854,
  title  = {$2$-Fault-Tolerant Strong Connectivity Oracles},
  author = {Loukas Georgiadis and Evangelos Kosinas and Daniel Tsokaktsis},
  journal= {arXiv preprint arXiv:2311.00854},
  year   = {2023}
}

Comments

Conference version to appear in the proceedings of ALENEX 2024

R2 v1 2026-06-28T13:09:05.729Z