Service Function Chain Mapping with Resource Fragmentation Avoidance
Zhikai Zhu, Hancheng Lu, Jian Li, Xiaoda Jiang · 2017
In the context of Service Function Chain (SFC), SFC Mapping (SFCM) problem has a decisive impact on the resource utilization efficiency of physical networks, where node resources and link resources are concerned. However, most existing work on the SFCM problem has not taken into account difference in the amount of node resources and link resources. This kind of difference might result in some nodes that cannot be mapped because of insufficient link resources around these nodes. This phenomenon is referred to as resource fragmentation. In this paper, we attempt to improve the resource utilization efficiency by reducing resource fragmentation in physical networks, where SFCM is performed. First and most importantly, we propose a metric called Resource Fragmentation Degree (RFD) to quantify resource fragmentation. The basic idea behind RFD is that the resource availability of a node is determined by the residual link resources around the node. Based on RFD, we formulate the SFCM problem with goal of minimizing resource fragmentation. Furthermore, we also propose an efficient online heuristic to find the optimal mapping strategy. Simulation results show that much more SFC requests can be accepted by reducing resource fragmentation in physical networks and the proposed algorithm achieves more than 25% higher acceptance ratio compared with existing algorithms.