Queuing analysis and optimization techniques for energy efficiency in packet networks

Energy efficiency in all aspects of human life has become a major concern, due to its significant environmental impact as well as its economic importance. Information and Communication Technology (ICT) plays a dual role in this; not only does it constitute a major consumer itself (estimated 2-10% of the global consumption), but is also expected to enable global energy efficiency through new technologies tightly dependent on networks (smart grid, smart homes, cloud computing etc.). To this purpose, this work studies the problem of energy efficiency in wired networks. As this subject has recently become very active in the research community, there is parallel research towards several research directions. In this work, the problem is being examined from its foundations and a solid analytical approach is presented. Specifically, a network model based on G-network queuing theory is built, which can incorporate all the important parameters of power consumption together with traditional performance metrics and routing control capability. This generalized model can be applied for any network case to build optimization algorithms and estimate the performance of different policies and network designs. Composite optimization goals functions are proposed, comprising both power consumption and performance metrics. A gradient descent optimization algorithm that can run in O(N3) time complexity is built thereof. Using power consumption characteristics of current and future equipment, several case studies are presented and the optimization results are evaluated. Moreover, a faster gradient-descent based heuristic and a decentralized algorithm are proposed. Apart from the routing control analysis, the case of a harsher energy saving solution, namely turning o the networking equipment, is also experimentally explored. Applying a tradeoff study on a laboratory testbed, implementation challenges are identified and conclusions significant for future work are drawn. Finally, a novel admission control mechanism is proposed and experimentally evaluated, which can monitor and manage the power consumption and performance of a network.Open Acces