GreenTE: Power-aware traffic engineering

Abstract

Abstract-Current network infrastructures exhibit poor power efficiency, running network devices at full capacity all the time regardless of the traffic demand and distribution over the network. Most research on router power management are at component level or link level, treating routers as isolated devices. A complementary approach is to facilitate power management at network level by routing traffic through different paths to adjust the workload on individual routers or links. Given the high path redundancy and low link utilization in today's large networks, this approach can potentially allow more network devices or components to go into power saving mode. This paper proposes an intra-domain traffic engineering mechanism, GreenTE, which maximizes the number of links that can be put into sleep under given performance constraints such as link utilization and packet delay. Using network topologies and traffic data from several widearea networks, our evaluation shows that GreenTE can reduce line-cards' power consumption by 27% to 42% under constraints that the maximum link utilization is below 50% and the network diameter remains the same as in shortest path routing.

Efficient heuristics for energy-aware routing in networks with bundled links

“…However, as argued in [16], this approach is not applicable to backbone networks that have short packet interval times. Nevertheless, their works [14,15] have inspired recent research on conserving energy in networks, including ours.…”

“…Thus, when the algorithm incorrectly removes a sub-optimal link, it will never backtrack to correct its mistake for the period. Other centralized optimization solutions on energy-aware TE, like ours, use shortest paths routing [13,16,21,22,24]. However, these studies do not consider links with bundled cables except [13].…”

“…There has been a handful of works on energy-aware Traffic Engineering (TE), some of which uses distributed optimization [17,18,19,20] while others utilize centralized optimization [13,16,21,22,23,24]. Vasic et al [17] present EATe, a technique that takes energy consumption into account while achieving the same traffic rates between source and destination nodes as energy oblivious approaches.…”

Greening the NetFPGA Reference Router

“…Many approaches have been proposed to make ICT systems more energy efficient [6,7]. The majority of these proposed approaches can be divided into two categories: power scaling and sleep mode [8][9][10].…”

Improvement of Steiner Tree Algorithm: Branch-Based Multi-Cast

“…This is because link-related resources, such as electricity used for the link-terminating ports and an Copyright c 2013 The Institute of Electronics, Information and Communication Engineers MPLS label for each link, can be conserved if the number of links used on a multicast tree is small. That is, if there is no traffic flowing in a link, the ports terminating the link can be set to sleep mode [18]. Once the MPLS protocol is used to establish the multicast tree, conserving MPLS labels is necessary [19].…”