The travelling miser problem

Various monitoring and performance evaluation tools generate considerable amount of low priority traffic. This information is not always needed in real time, and thus could often be delayed by the network without hurting functionality. This paper proposes a new framework to handle this low priority, but resource consuming traffic in such a way that it will incur a minimal interference with the higher priority traffic, and thus improve the network goodput. The key idea is to allow the network nodes to delay data by locally storing it. This can be done, for example, in the active network paradigm. We show that the active network paradigm can improve the network's goodput dramatically even if a very simple scheduling algorithm is used. To obtain minimal cost schedules we define an optimization problem that we call the travelling miser problem. We study primarily the on-line variant of this problem, which is of greater practical interest. For this problem we develop an enhanced scheduling strategy, study its characteristics in a restricted case, and evaluate its performance through a rigorous simulation study.