Self-Adaptation in Next-Generation Internet Networks: How to React to Traffic Changes While Respecting QoS?

Next-generation network infrastructure should support different services and several levels of quality of service (QoS) and resilience. The main requirements for such multiservice networks are flexibility, effective utilization of network resources, and ability to react to traffic demand changes with time. These requisites lead to the fact that next-generation networks should, to some extend, be "self-adapting", to cope with the need of promptly reacting to traffic demand changes, in an effective way, while taking into account the different requirements in terms of QoS. This paper sketches the reference network scenario and discusses the subject of self-adapting networks, attempting to focus on the key issues that need to be solved to realize future infrastructures. As a relevant example, this paper briefly reports a system developed by the authors' laboratory that implements a traffic engineering solution able to fulfil the mentioned requirements. This system could be regarded as a first concrete step toward the realization of self-adapting networks. In addition, some relevant results that were obtained either by simulation or by experiments on a real test bed are reported to discuss the main characteristics of such a system and assess the feasibility of the concept. Finally, the paper reviews the main hot issues that need to be addressed, in the authors' opinion, by the international research community

[1]  Dave Katz,et al.  Traffic Engineering (TE) Extensions to OSPF Version 2 , 2003, RFC.

[2]  Marco Listanti,et al.  Architectural and technological issues for future optical Internet networks , 2000, IEEE Commun. Mag..

[3]  Daniel O. Awduche,et al.  Requirements for Traffic Engineering Over MPLS , 1999, RFC.

[4]  Alessio Botta,et al.  Internet like control for MPLS based traffic engineering: performance evaluation , 2005, Perform. Evaluation.

[5]  Xipeng Xiao,et al.  Internet QoS: a big picture , 1999, IEEE Netw..

[6]  Ajay D. Kshemkalyani,et al.  Connection preemption: issues, algorithms, and a simulation study , 1997, Proceedings of INFOCOM '97.

[7]  Lionel M. Ni,et al.  Traffic engineering with MPLS in the Internet , 2000, IEEE Netw..

[8]  Erol Gelenbe,et al.  Autonomous smart routing for network QoS , 2004 .

[9]  Paola Iovanna,et al.  Performance analysis of a traffic engineering solution for multilayer networks based on the GMPLS paradigm , 2004, IEEE Journal on Selected Areas in Communications.

[10]  Paola Iovanna,et al.  A traffic engineering system for multilayer networks based on the GMPLS paradigm , 2003, IEEE Netw..

[11]  Gianpaolo Oriolo,et al.  A multilayer solution for path provisioning in new-generation optical/MPLS networks , 2003 .

[12]  Vijay Srinivasan,et al.  RSVP-TE: Extensions to RSVP for LSP Tunnels , 2001, RFC.

[13]  Ayan Banerjee,et al.  Generalized multiprotocol label switching: an overview of routing and management enhancements , 2001, IEEE Commun. Mag..

[14]  Roberto Sabella,et al.  An Innovative Solution for Dynamic Bandwidth Engineering in IP/MPLS Networks with QoS Support , 2004, Photonic Network Communications.

[15]  Erol Gelenbe,et al.  Self-aware networks and QoS , 2004, Proceedings of the IEEE.

[16]  Gianpaolo Oriolo,et al.  Strategy for Dynamic Routing and Grooming of Data Flows into Lightpaths in New Generation Network Based on the GMPLS Paradigm , 2004, Photonic Network Communications.