Centralized, Parallel, and Distributed Multi-Source Shortest Paths via Hopsets and Rectangular Matrix Multiplication
Abstract
Consider an undirected weighted graph . We study the problem of computing -approximate shortest paths for , for a subset of sources, for some . We devise a significantly improved algorithm for this problem in the entire range of parameter , in both the classical centralized and the parallel (PRAM) models of computation, and in a wide range of in the distributed (Congested Clique) model. Specifically, our centralized algorithm for this problem requires time , where is the time required to multiply an matrix by an one. Our PRAM algorithm has polylogarithmic time , and its work complexity is , for any arbitrarily small constant . In particular, for , our centralized algorithm computes -approximate shortest paths in time. Our PRAM polylogarithmic-time algorithm has work complexity , for any arbitrarily small constant . Previously existing solutions either require centralized time/parallel work of or provide much weaker approximation guarantees. In the Congested Clique model, our algorithm solves the problem in polylogarithmic time for sources, for , while previous state-of-the-art algorithms did so only for . Moreover, it improves previous bounds for all . For unweighted graphs, the running time is improved further to .
Cite
@article{arxiv.2004.07572,
title = {Centralized, Parallel, and Distributed Multi-Source Shortest Paths via Hopsets and Rectangular Matrix Multiplication},
author = {Michael Elkin and Ofer Neiman},
journal= {arXiv preprint arXiv:2004.07572},
year = {2021}
}