Technical Note: \cal Q-Learning
Machine Learning
The dynamics of reinforcement learning in cooperative multiagent systems
AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Proceedings of the 2003 conference on Applications, technologies, architectures, and protocols for computer communications
Energy-aware traffic engineering
Proceedings of the 1st International Conference on Energy-Efficient Computing and Networking
Reducing power consumption in backbone networks
ICC'09 Proceedings of the 2009 IEEE international conference on Communications
The green-game: striking a balance between QoS and energy saving
Proceedings of the 23rd International Teletraffic Congress
Energy management in IP traffic engineering with Shortest Path routing
WOWMOM '11 Proceedings of the 2011 IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks
Minimizing ISP network energy cost: formulation and solutions
IEEE/ACM Transactions on Networking (TON)
Energy-aware IP traffic engineering with shortest path routing
Computer Networks: The International Journal of Computer and Telecommunications Networking
Sleep modes effectiveness in backbone networks with limited configurations
Computer Networks: The International Journal of Computer and Telecommunications Networking
| Hi-index | 0.00 |
In this work, we face the problem of reducing the power consumption of Internet backbone networks. We propose a novel algorithm, called GRiDA, to selectively switch off links in an Internet Service Provider IP-based network to reduce the system energy consumption. Differently from approaches that have been proposed in the literature, our solution is completely distributed and thus not requiring any centralized oracle. It leverages link-state protocol like OSPF to share a limited amount of information, and to reduce the problem complexity. Another key feature of GRiDA is that it does not require the knowledge of the actual and past/future traffic matrices, being able to run in real-time, where this information would not be available. Results, obtained on realistic case studies, show that GRiDA achieves performance comparable to several existing centralized algorithms, guaranteeing energy savings up to 50%.