Recent works analyze the energy saving improvements of greened networks. The main idea is to define algorithms able to put some network links or devices into sleep mode for saving the used energy. This can achieved by adopting power-aware traffic engineering to route traffic towards a subset of links or to consider topology-based and traffic-based pruning mechanisms. These approaches are very promising as for the energy saving and the overall network performance remain satisfactory, however the question is: are the per-flow traffic performance still guaranteed? How does TCP perform in such a greened network? In this paper we analyze the performance of different versions of TCP in a network where energy savings are achieved by the use of one of the algorithm, namely ESTOP, defined to prune the IP network on the basis of network topological characteristics. The performance results show that the behavior of that TCP versions depending on the round trip time, namely Reno and Cubic, are affected by the link pruning in an expected way. This may allow to use these pruning techniques with a suitable control on the per-flow performance. The tradeoff between energy saving and TCP through-put loss is analyzed.
[1]
Alexander Afanasyev,et al.
Host-to-Host Congestion Control for TCP
,
2010,
IEEE Communications Surveys & Tutorials.
[2]
Jitender S. Deogun,et al.
TCP Congestion Avoidance Algorithm Identification
,
2011,
2011 31st International Conference on Distributed Computing Systems.
[3]
Marco Listanti,et al.
Distributed and Adaptive Interface Switch Off for Internet Energy Saving
,
2011,
2011 Proceedings of 20th International Conference on Computer Communications and Networks (ICCCN).
[4]
Alessandro Carrega,et al.
Green network technologies and the art of trading-off
,
2011,
2011 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS).
[5]
Francesca Cuomo,et al.
Network pruning for energy saving in the Internet
,
2012,
Comput. Networks.
[6]
Michal Pióro,et al.
SNDlib 1.0—Survivable Network Design Library
,
2010,
Networks.
[7]
Marco Mellia,et al.
Distributed algorithms for green IP networks
,
2012,
2012 Proceedings IEEE INFOCOM Workshops.