Andrés López Martínez

Parallel Minimum Cuts

Abstract

This thesis considers the minimum cut problem in undirected, weighted graphs. We present a simple randomized CREW PRAM algorithm to find the minimum cut in a graph \(G\) with \(n\) nodes and \(m\) edges, based on Karger’s tree-packing approach. It has near-linear work \(O(m \log^2 n + n \log^6 n)\) and low depth \(O(\log^3 n)\), and returns the correct result with high probability. This is the first improvement to the best previous \(O(m \log^4 n)\) work and \(O(\log^3 n)\) depth CREW PRAM min-cut algorithm by Geissmann and Gianinazzi. For his randomized near-linear min-cut algorithm Karger used a connection between minimum cuts and maximum packings of spanning trees to reduce the minimum cut problem into two subproblems: (i) compute an approximate maximum tree packing \(\mathcal{S}\), and (ii) find the minimum cut of the graph \(G\) that cuts at most two edges from some tree in \(\mathcal{S}\)—this is referred to as the 2-respecting min-cut problem. To achieve our main result, we give parallel algorithms for both subproblems. More precisely, we present the following.

In addition, we develop the following independent results: