It's Good to Relax: Fast Profit Approximation for Virtual Networks with Latency Constraints

Abstract

This paper proposes a new approximation algorithm for the offline Virtual Network Embedding Problem (VNEP) with latency constraints. Given is a set of virtual networks with computational demands on nodes and bandwidth demands together with latency bounds on the edges. The VNEP's task is to feasibly embed a subset of virtual networks on a shared physical infrastructure, e.g., a data center, while maximizing the attained profit. In contrast to existing works, our approximation algorithm AFlex allows for (slight) violations of the latency constraints in order to greatly lower the runtime. To obtain this result, we use a reduction to the Restricted Shortest Path Problem (RSP) and leverage a classic result by Goel et al. We complement our formal analysis with an extensive simulation study demonstrating the computational benefits of our approach empirically. Notably, our results generalize to any other additive edge metric besides latency, including loss probability.