Performance bounds of rate-adaptation schemes for energy-efficient routers

Abstract

To maximize the energy efficiency of packet networks, energy use in network equipment should scale rigorously with the traffic load. Rate adaptation is gaining popularity as a promising framework for achieving energy proportionality in the data-path components of routers and switches, but a thorough characterization of its energy-saving capabilities and impact on network performance is still lacking. In this paper, we present novel rate-adaptation schemes that are easy to implement, amenable to incremental deployment in packet switches and routers, and whose negative effect on end-toend packet delays is bounded. We derive tight lower and upper bounds on energy consumption that we use to further refine the design of our schemes. Our analysis indicates that large-scale deployments of rate adaptation can enable substantial energy savings, as large gains can be attained in individual data-path devices under parametric configurations that do not compromise the end-to-end delay performance of the network.

Minimizing the monetary penalty and energy cost of server migration service

“…The assumption of the line‐card energy consumption depending on the traffic load of the input link connected to the line‐card is justified, because a module for processing layers 1 to 3 functionality in a line‐card is mainly devoted to processing the incoming traffic load (packets) on the input link § The assumption of the energy‐efficient line‐card adjusting its energy consumption based on its traffic load is justified, because the energy‐efficient line‐card has rate adaptation capability, that is, the line‐card adjusts its energy consumption based on its traffic load by dynamically adjusting the voltage and frequency of its module for processing layers 1 to 3 functionality 41 …”

“…The assumption of the energy-efficient line-card adjusting its energy consumption based on its traffic load is justified, because the energy-efficient line-card has rate adaptation capability, that is, the line-card adjusts its energy consumption based on its traffic load by dynamically adjusting the voltage and frequency of its module for processing layers 1 to 3 functionality. 41 For a WP, we assume the following energy consumption model and the energy cost model.…”

Rate-Adaptive Scheduling Policies for Network Stability and Energy Efficiency

“…It exhibits aspects of a convex (superadditive) function due to the , but also exhibits aspects of a subadditive function due to the term. Although hardware support for power management for network elements is in its initial stage, packet-timescale rate adaptation can be realized by adjusting the clock frequency and supply voltage of data-path hardware components to locally maintained workload indicators such as queue lengths and traffic arrival rates [10], [9], [13]. The timescale of rate change and the resulting energy consumption depend on the underlying integrated circuit technology, e.g., CMOS.…”

Traffic Estimation for Dynamic Capacity Adaptation in Load Adaptive Network Operation Regimes

“…Among them a prominent opportunity for improving the energy efficiency of networks is load-adaptive operation where network capacity follows traffic demands to a certain extent using different technical approaches (e.g. (Antonakopoulos et al, 2010;Francini and Stiliadis, 2010;Puype et al, 2011;Lange and Gladisch, 2011)). This is in contrast to the prevalent network design and building practice -where network capacity is above all provided based on the expected peak traffic plus a capacity reserve.…”