Descending-Order Clique-Based Task Scheduling for Active Measurements

Abstract

Network active measurement is a required process in high performance networks for efficient QoS provisioning. The contention of the active measurement processes in a network can jeopardize measurement and overload network and computation resources. In this paper, we propose a novel scheduling algorithm to resolve contention of measurement tasks. Our proposed scheme is based on graph-coloring theory and uses a descending order of a clique (a set of contending nodes) in a graph. We evaluate our proposed scheme under different measurement task assignment scenarios. The simulation results demonstrate that our proposed scheme produces effective contention resolution and decreases the number of assigned colors.