Skip to Main content Skip to Navigation
Conference papers

Frugal Topologies for Saving Energy in IP Networks

Abstract : Recent years have seen the advent of energy conservation as a key engineering and scientific challenge of our time. At the same time, most IP networks are typically provisioned to 30%-40% average utilization, leading to a significant waste of energy. Current approaches for creating more frugal networks rely on instantaneous and global knowledge of the traffic matrix and network congestion levels - a requirement that can be impractical for many network operators. We introduce a new traffic-agnostic metric for quantifying the quality of a frugal topology, the Adequacy Index (ADI).We show that the problem of minimizing the power consumption of a network subject to a given ADI threshold is NP-hard and present two polynomial time heuristics - ABStAIn and CuTBAck. We perform extensive simulations using topologies and traffic matrices from 3 real networks. Our results show that ABStAIn and CuTBAck are as effective as an exponential time traffic based solution at creating frugal topologies and outperform a state of the art polynomial time traffic based solution by about 80%. Furthermore, the median link utilization observed with ABStAIn and CuTBAck is similar to that with traffic based solutions, with the maximum link utilization never exceeding 80%.
Complete list of metadata

Cited literature [24 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01334682
Contributor : Open Archive Toulouse Archive Ouverte (oatao) Connect in order to contact the contributor
Submitted on : Tuesday, June 21, 2016 - 11:02:09 AM
Last modification on : Tuesday, October 19, 2021 - 2:24:23 PM
Long-term archiving on: : Thursday, September 22, 2016 - 11:05:57 AM

File

hussein_15416.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01334682, version 1
  • OATAO : 15416

Citation

Mohammed Hussein, Gentian Jakllari, Béatrice Paillassa. Frugal Topologies for Saving Energy in IP Networks. 40th IEEE Conference on Local Computer Networks (LCN 2015), Oct 2015, Clearwater Beach, Florida, United States. pp. 303-311. ⟨hal-01334682⟩

Share

Metrics

Record views

198

Files downloads

391