Skip to main content

Research Repository

Advanced Search

Vicinity-based replica finding in named data networking

Suwannasa, Atthapol; Broadbent, Matthew; Mauthe, Andreas


Atthapol Suwannasa

Matthew Broadbent

Andreas Mauthe


In Named Data Networking (NDN) architectures, a content object is located according to the content's identifier and can be retrieved from all nodes that hold a replica of the content. The default forwarding strategy of NDN is to forward an Interest packet along the default path from the requester to the server to find a content object according to its name prefix. However, the best path may not be the default path, since content might also be located nearby. Hence, the default strategy could result in a sub-optimal delivery efficiency. To address this issue we introduce a vicinity-based replica finding scheme. This is based on the observation that content objects might be requested several times. Therefore, replicas can be often cached within a particular neighbourhood and thus it might be efficient to specifically look for them in order to improve the content delivery performance. Within this paper, we evaluate the optimal size of the vicinity within which content should be located (i.e. the distance between the requester and its neighbours that are considered within the content search). We also compare the proposed scheme with the default NDN forwarding strategy with respect to replica finding efficiency and network overhead. Using the proposed scheme, we demonstrate that the replica finding mechanism reduces the delivery time effectively with acceptable overhead costs.

Presentation Conference Type Conference Paper (Published)
Conference Name 2020 International Conference on Information Networking (ICOIN)
Start Date Jan 7, 2020
End Date Jan 10, 2020
Online Publication Date Mar 2, 2020
Publication Date 2020
Deposit Date Mar 9, 2022
Publisher Institute of Electrical and Electronics Engineers
Pages 146-151
Series ISSN 1976-7684
Book Title 2020 International Conference on Information Networking (ICOIN)
Public URL