Radient: Scalable, memory efficient name lookup algorithm for named data networking Divya Saxena n , Vaskar Raychoudhury Department of Computer Science and Engineering, Indian Institute of Technology (IIT), Roorkee, India article info Article history: Received 29 May 2015 Received in revised form 9 October 2015 Accepted 25 December 2015 Available online 22 January 2016 Keywords: Named data networking (NDN) Name encoding Pending interest table (PIT) abstract Named Data Networking (NDN) aims to discard the existing host-centric networking paradigm just to replace it with a more practical Content-Centric Networking (CCN) paradigm. CCN allows users to fetch and distribute contents directly using their names. NDN router stores all incoming content requests ('/'-delimited string components) in the Pending Interest Table (PIT) until they are satisfied. Multiple requests for the same content are merged in a single PIT entry and when the requested content is available, it is forwarded simultaneously to all the requesters. Although NDN has several benefits over the existing IP-based network, replacing IP addresses with names increases memory consumption and lookup cost. One possible way to restrict memory usage is to use name encoding, i.e., to encode identical components of a name with a unique integer. In this paper, we proposed a novel memory efficient name encoding scheme (called, Radient) for PIT and evaluated it extensively. Our results show that the Radient scheme can reduce memory consumption by 35.45% compared to the ENPT for 29 million names. & 2016 Elsevier Ltd. All rights reserved. 1. Introduction Named Data Networking (NDN) is the most promising candi- date among the content-based future Internet design paradigms. In the current TCP/IP based model, packets use IP address to identify and fetch data from specific end-hosts. NDN, on the other hand, uses unique names to pull content irrespective of their hosting entity (Jaffri et al., 2013; Hu et al., 2011; Pan et al., 2011; Papadimitriou et al., 2010). As per the NDN design specification (Zhang et al., 2010), content names are not transparent to the network (i.e., NDN routers never try to interpret the names, but simply forward), and are used to make routing and forwarding decisions. In NDN, a content name may contain several compo- nents (string of arbitrary length) arranged in a hierarchy (delim- ited by '/') whereas IP uses fixed length (either 32 bit or 64 bit) addresses. When an NDN router receives a request (Interest packet) for a specific content, it is passed to the set of functional data struc- tures: Content Store (CS), Pending Interest Table (PIT) and For- warding Information Base (FIB). The CS caches previously accessed contents in order to satisfy matching future requests. If a reques- ted content is not available in the CS, the content name is checked in the PIT and if content name does not exists, then the requesters interface is appended to the list of interfaces waiting to receive the content. The corresponding Interest packet is passed to the FIB and a new PIT entry is created. A longest prefix of the content name is calculated and Interest is forwarded to the outgoing interface(s) of FIB's longest prefix matching entry. Though accessing content using application name solves many problems associated with current IP based networking, yet NDN is facing two major challenges. Firstly, storing variable length names consume higher memory in comparison to an IP address. E.g., NDN name may have on average 5–6 components. If, each component has 5 characters, the total memory consumed is 25 bytes for storing an NDN name. Thus, NDN routers can quickly get depleted of available memory space compared to IP routers. Secondly, each time a content request or a corresponding reply arrives at the NDN router, the entire PIT is lookup using an exact matching (character- to-character matching) technique to check whether an entry for the content name already exists or not. A simple lookup scheme incurs time proportional to the number of PIT entries as well as the length of individual names. As evident from the aforementioned shortcomings, we require an efficient and dynamic data structure for storing names in the PIT of NDN routers. The data structure should support less mem- ory consumption, low computational complexity, and fast name lookup. Some conventional data structures like, Hash Table, Component Character Trie (CCT), and Name Prefix Trie (NPT) are used in exact matching algorithms. Using Hash Table, either the whole content name or its components are mapped into associa- tive array using a hash function. This structure incurs less memory Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/jnca Journal of Network and Computer Applications http://dx.doi.org/10.1016/j.jnca.2015.12.009 1084-8045/& 2016 Elsevier Ltd. All rights reserved. n Corresponding author. Tel.: þ91 7351771180. E-mail addresses: divya.saxena.2015@ieee.org (D. Saxena), vaskar@ieee.org (V. Raychoudhury). Journal of Network and Computer Applications 63 (2016) 1–13