Approximating k-Edge-Connected Spanning Subgraphs via a Near-Linear Time LP Solver

In the $k$-edge-connected spanning subgraph ($k$ECSS) problem, our goal is to compute a minimum-cost sub-network that is resilient against up to $k$ link failures: Given an $n$-node $m$-edge graph with a cost function on the edges, our goal is to compute a minimum-cost $k$-edge-connected spanning subgraph. This NP-hard problem generalizes the minimum spanning tree problem and is the "uniform case" of a much broader class of survival network design problems (SNDP). A factor of two has remained the best approximation ratio for polynomial-time algorithms for the whole class of SNDP, even for a special case of $2$ECSS. The fastest $2$-approximation algorithm is however rather slow, taking $O(mn k)$ time [Khuller, Vishkin, STOC'92]. A faster time complexity of $O(n^2)$ can be obtained, but with a higher approximation guarantee of $(2k-1)$ [Gabow, Goemans, Williamson, IPCO'93]. Our main contribution is an algorithm that $(1+ε)$-approximates the optimal fractional solution in $\tilde O(m/ε^2)$ time (independent of $k$), which can be turned into a $(2+ε)$ approximation algorithm that runs in time $\tilde O\left(\frac{m}{ε^2} + \frac{k^2n^{1.5}}{ε^2}\right)$ for (integral) $k$ECSS; this improves the running time of the aforementioned results while keeping the approximation ratio arbitrarily close to a factor of two.