A producer-centric caching strategy for named data networking

Abstract

With growing number of users and content on Internet, the existing Internet architecture needs to evolve to be more efficient to find and carry content, and Named Data Networking (NDN) is a new Internet architecture aiming to achieve this. Named Data Networking focuses on data directly instead of considering servers and their locations to retrieve data. It adopts the paradigm of requesting and retrieving data with name instead of delivering a packet to a specific destination address. In NDN, caching of data in intermediate network nodes can help moving data closer to the consumers and in this way can be very e ective in reducing application response time and network bandwidth consumption. For this, proper cache placement and replacement strategies are needed. In this thesis, we propose a producer-centric in-network caching strategy, called PCDC, to efficiently and e ectively cache data in the intermediate nodes of a named data network. In our scheme, intermediate nodes that are at the intersections of multiple consumer paths are selected as caching points to cache requested data. We propose the necessary mechanisms and algorithms to select such caching points, considering the practical constrains of NDN protocols. We simulated our scheme in ndnSIM/ns3 and did extensive simulation experiments to evaluate its performance. The simulation results show that our scheme is feasible and practical to apply and is efficient in terms of increasing packet diversity and average hit-ratio, and decreasing application response time and network bandwidth consumption.