Optimising data placement and traffic routing for energy saving in Backbone Networks

Abstract

The energy consumption of network elements has become a big concern due to the exponential traffic growth and the rapid expansion of communication infrastructures. To deal with this problem, we propose a new approach called Backbone network Energy Saving based on Traffic engineering (BEST), which reduces the power consumption of network elements at the backbone level through jointly optimising data placement and traffic routing. Based on analysis on traffic characteristics, BEST firstly optimises the placement of data services such that the pairwise traffic demands can be better coordinated with the pairwise network costs, in order to minimise the traffic burden imposed on the network elements. Then, BEST optimises the routing of traffic flows and tries to find the minimum‐power network subset that must remain active to fulfill the traffic requirements. Efficient heuristics are given by BEST to find an admissible solution when the problem size is very large. The simulation results illustrate the efficacy and efficiency of BEST in energy conservation on backbone networks. Copyright © 2013 John Wiley & Sons, Ltd.

Joint content placement and lightpath routing and spectrum assignment in CDNs over elastic optical network scenarios

“…Moreover, there are some papers that address the problems of designing CDNs (e.g., [28], [29]) and content placement in CDNs (or equivalent problems), e.g., [30], [31], [32], [33], [34], [35]. However, these papers do not address routing of content requests or they assume a very simple routing model.…”

Power‐aware routing in networks with quality of services constraints

“…A number of works use distributed optimisation [13][14][15][16] while others utilise centralised optimisation [9,11,[17][18][19][20][21][22][23][24][25]. A distributed, low-complexity component on routers is proposed in [13] to monitor link load state.…”

“…Their method selects the least used link as a candidate link. The work in , called BEST, uses a similar idea, that is, it orders network elements based on their occurrence in routing paths and places those less used resources into sleep.…”

An energy‐aware engineered control plane for wavelength‐routed networks

“…The work presented in exploits with a dynamic multi‐stage approach the minimisation of both energy consumption and carbon footprint in WDM networks providing adaptive sub‐wavelength routing facilities. In , energy savings in backbone networks is achieved through joint optimisation of data placement and traffic routing. Most of the aforementioned approaches are characterised by multi‐stage complex decision mechanisms or a quite limited dynamism and hence are not easily and seamlessly applicable in a fully adaptive online scenario, often using power containment techniques based on switching down inactive elements.…”