A Probabilistic Interest Packet Aggregation for Content-Centric Networking
Yuichi Yasuda, Ryo Nakamura, Hiroyuki Ohsaki · 2018
Recently, Content-Centric Networking (CCN) has been extensively studied by networking researchers as a promising network architecture to realize an information-centric network. When a CCN router receives multiple Interest packets requesting identical content, it can aggregate those packets into a single Interest packet in order to reduce the amount of redundant Interest and Data packet transmission. In the literature, it is known that Interest packet aggregation significantly affects the performance of a CCN network. In this paper, we propose a method called Probabilistic Interest Packet Aggregation (PIPA). We investigate the fundamental characteristics of PIPA - in particular, the relation between its Interest packet aggregation probability and the average chunk delivery delay - through both simulation experiments and mathematical analyses. Our findings include that by appropriately aggregating Interest packets at a CCN router using PIPA, the average chunk delivery delay can be reduced by approximately 15% compared with the case without PIPA.