Towards Popularity-Based Caching in Content Centric Networks

César Bernardini 1 Thomas Silverston 1 Festor Olivier 1
1 MADYNES - Management of dynamic networks and services
Inria Nancy - Grand Est, LORIA - NSS - Department of Networks, Systems and Services
Abstract : Today's Internet was created aiming to share resources. The chosen model was therefore a conversation mechanism between two machines: the messages are exchanged between two host according to source and destination addresses of IP datagrams. This led to a model with strong coupling between where and what, even when most of the users value the Internet for what content it contains. As well location dependence makes configuration and implementation of network services more complex. Alternative architectures such as CDN and P2P networks provide high availability and improve performance but they are not-supported natively by the current Internet. In order to solve these issues, Information-Centric Network architectures (ICN) have been proposed such as PSIRP, NetInf, or CCN. These architectures are thought as clean-state designs for the Future Internet. Content Centric Networking (CCN) is a communication architecture based on named data where packet address name content, not location. The host notion does not exist anymore. The communication paradigm relies on two primitives: Interest and Data. A consumer requests content by broadcasting its interest all over the network; any node hearing the request and having the data can issue a response with a data message. In addition, CCN offers routing-by-name, caching, integrity checks based on signatures, encryption, multicast support, flow-control at each hop, protection against Denial of Service attacks. Caching consists in storing data messages at a node. In CCN, data messages are cached at every intermediate nodes from a source to a destination. A content is therefore replicated and is available at a closer distance for future requests. An important feature for CCN is to manage the cache of nodes by defining replacement policy, cache policy and cache strategy. Most of the research about information centric networks, have been carried out on CCN. For instance, Kurose et al. show that cache replacement policies can be grouped in the same equivalence class. Rossi et al. study cache sizes and compare several caching strategies. They also argue that popularity of content have a strong impact on CCN performances[4]. In this extended abstract, we propose a new caching strategy based on popularity.
Type de document :
Communication dans un congrès
RESCOM 2012, Jun 2012, Les Vosges, France. 2012
Liste complète des métadonnées

Littérature citée [7 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00747611
Contributeur : César Bernardini <>
Soumis le : mercredi 31 octobre 2012 - 16:36:02
Dernière modification le : jeudi 11 janvier 2018 - 06:25:23
Document(s) archivé(s) le : vendredi 1 février 2013 - 03:40:06

Fichier

bare_conf.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00747611, version 1

Collections

Citation

César Bernardini, Thomas Silverston, Festor Olivier. Towards Popularity-Based Caching in Content Centric Networks. RESCOM 2012, Jun 2012, Les Vosges, France. 2012. 〈hal-00747611〉

Partager

Métriques

Consultations de la notice

468

Téléchargements de fichiers

562