Knowledge Dissemination for Autonomic Network

Sami Souihi (1) , Abdelhamid Mellouk (1)

1 CIR

Autonomic computing is a new paradigm inspired by the biological world. It allows networks to self-organize control decisions. To ensure an efficient self-organization, an autonomic network must define a new distributed and decentralized plane containing all the network knowledge, called "knowledge plane". In this paper, we propose a new knowledge dissemination mechanisms in order to maintain this knowledge plane. The developed approach is composed by three modules to address scalability and interoperability issues: (1) cluster the network nodes and elect a leader for each cluster to contain the knowledge, (2) use an overlay network to disseminate knowledge between leaders for interoperability concerns and, (3)disseminate only useful knowledge to nodes in each cluster to reduce the amount of knowledge to be distributed. To evaluate our approach, we construct a knowledge plan based on QoE measurement. Numerical results obtained for IPTV different traffics level merged on operators network show that our approach improves clearly performances regarding load balancing charges.

Mots clés: IPTV; computer networks; fault tolerant computing; open systems; pattern clustering; QoE measurement; autonomic computing; autonomic network; interoperability; knowledge dissemination mechanisms; knowledge plane; network knowledge; overlay network; self-organize control decisions; Clustering algorithms; Complexity theory; Error analysis; Knowledge engineering; Peer to peer computing; Scalability; Streaming media

Domaines: Informatique

Sami Souihi, Abdelhamid Mellouk. Knowledge Dissemination for Autonomic Network. 2011 IEEE International Conference on Communications (ICC), Jun 2011, Kyoto, Japan. pp.1--5, ⟨10.1109/icc.2011.5962880⟩. ⟨hal-00917830⟩