2018/07/20 by Amir Varasteh, Marilet De Andrade, Varasteh, Amir +7
Computer Science · Engineering · #Advanced Optical Network Technologies #FOS: Computer and information sciences #Interconnection Networks and Systems #Networking and Internet Architecture (cs.NI) #Software-Defined Networks and 5G
paper · pdf · doi:10.48550/arxiv.1807.07876
openalex publication_date 2018/07/20 · openalex created_date 2022/08/04 · openalex updated_date 2026/07/28
The Network Function Virtualization (NFV) is very promising for efficient\nprovisioning of network services and is attracting a lot of attention. NFV can\nbe implemented in commercial off-the-shelf servers or Physical Machines (PMs),\nand many network services can be offered as a sequence of Virtual Network\nFunctions (VNFs), known as VNF chains. Furthermore, many existing network\ndevices (e.g., switches) and collocated PMs are underutilized or\nover-provisioned, resulting in low power-efficiency. In order to achieve more\nenergy efficient systems, this work aims at designing the placement of VNFs\nsuch that the total power consumption in network nodes and PMs is minimized,\nwhile meeting the delay and capacity requirements of the foreseen demands.\nBased on existing switch and PM power models, we propose a Integer Linear\nProgramming (ILP) formulation to find the optimal solution. We also propose a\nheuristic based on the concept of Blocking Islands (BI), and a baseline\nheuristic based on the Betweenness Centrality (BC) property of the graph. Both\nheuristics and the ILP solutions have been compared in terms of total power\nconsumption, delay, demands acceptance rate, and computation time. Our\nsimulation results suggest that BI-based heuristic is superior compared with\nthe BC-based heuristic, and very close to the optimal solution obtained from\nthe ILP in terms of total power consumption and demands acceptance rate.\nCompared to the ILP, the proposed BI-based heuristic is significantly faster\nand results in 22% lower end-to-end delay, with a penalty of consuming 6% more\npower in average.\n