Effective controller placement in controller-based Named Data Networks - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

Effective controller placement in controller-based Named Data Networks

Abstract

Named Data Networking (NDN) has been recently touted as one of the most appealing future Internet architectures. One prominent feature in NDN networks is the in-network caching that allows routers to store content in their cache and directly serve it to end-users. In this context, efficiently forwarding content requests to the closest router having the desired content has been a daunting challenge that still needs to be addressed. To tackle this challenge and inspired by software defined networking paradigm, we recently proposed a controller-based NDN caching and forwarding scheme where multiple controllers cooperate to efficiently handle request forwarding in the network [1]. In this paper, we further explore this solution by addressing two related challenges: 1) determining the optimal number of controllers able to handle the network traffic; 2) find an optimal placement for the controllers that minimizes the controller load and the inter-controller latency. We first formulate the controller placement problem as an integer linear program and then study the performance of two potential clustering-based solutions. Extensive simulations using real network topologies show that these solutions can provide near-optimal controller locations which, in turn, improve the performance of controller-based NDN forwarding schemes in terms of data download latency and throughput.
No file

Dates and versions

hal-02099050 , version 1 (13-04-2019)

Identifiers

Cite

Narjes Aloulou, Mouna Ayari, Mohamed Faten Zhani, Leila Saidane, Guy Pujolle. Effective controller placement in controller-based Named Data Networks. 2017 International Conference on Computing, Networking and Communications (ICNC), Jan 2017, Santa Clara, CA, United States. pp.249-254, ⟨10.1109/ICCNC.2017.7876134⟩. ⟨hal-02099050⟩
47 View
0 Download

Altmetric

Share

Gmail Facebook X LinkedIn More