The characteristics of traffic in core networks have changed in recent years, with the data flows among network nodes now showing large volumes but also large fluctuations during the day. In such a dynamic environment, the static allocation of network resources by making reservations at the peak rates is wasteful, both in terms of energy consumption and resource efficiency. Periodic reconfiguration of network resources is necessary for the full exploitation of an elastic optical network’s potential. In this paper, we present a resource reconfiguration scheme for an IP over elastic optical core network. The cornerstone of the proposed scheme is a novel traffic prediction mechanism based on reinforcement learning. Using the forecasting mechanism, a heuristic routing and spectrum assignment algorithm was designed to efficiently and fairly allocate network resources periodically. Simulation results show both a reduction of the bandwidth blocking probability and an increase of the fairness index regarding the network resources allocation over corresponding algorithms in the literature.

To date, the main limitation to the development of telecom networks mainly depended on two factors, namely the cost of the network equipment and transmission. However, nowadays, the enormous spread and growth of the Internet has resulted in a significant increase in the energy consumption of the network equipment. Particularly, the rise in Internet traffic has led to both the increase of the amount of required equipment and the augmentation of energy consumption of such network devices. Consequently, CapEx and OpEx of modern telecommunication networks render the market penetration of such deployment laborious. This surge in power consumption may ultimately lead to an impediment to this continued growth. Since climate change and environmental protection are also of rising concern, many research efforts are being made to raise awareness of environmental issues and reduce the energy footprint of telecommunication networks. In this context, this study targets the reduction of the energy consumption of an IP over WDM backbone network. Already known lightpath bypass and traffic grooming strategies have been used. Although previously proposed schemes based on lightpath bypass approaches and traffic grooming capabilities have been implemented and consequently are able to reduce the overall power consumption of an optical core network, they do not bear in mind, first, the number of the established lightpaths in the virtual topology and second, the exploitation of the residual capacity of the existing lightpaths. This study overcomes the above drawbacks by introducing three novel heuristic schemes which manage to significantly lower the number of the established lightpaths in the virtual topology and take advantage of the remaining capacity of the optical paths by exploiting the lightpaths’ resources in a more efficient way. These power-aware heuristics will focus on energy efficiency issues related to the planning and the operation phase of a WDM optical network. Extensive simulation results indicate a promising performance improvement in terms of power consumption when these proposed heuristics are compared to similar, already developed, techniques for energy saving.

Communication networks for healthcare environments support various application types and should be capable of providing them with the required Quality of Service (QoS) in terms of reliable data delivery, considerable data rate and low latency. The introduction of Tactile Internet (TI) healthcare applications is expected to upgrade the level of provided services to patients. The requirements of these applications are very stringent and a major requirement is that the existing network infrastructure must support them. Passive Optical Networks (PONs) have been proposed as an ideal candidate to support such demanding environments. One of the major challenges of such networks, which significantly affects the provided QoS, is the effective resource allocation in both the time and the wavelength domains. The existing resource allocation algorithms for PONs were designed without considering the stringent requirements of TI applications thus making PON support for TI inefficient. In this paper, a new double per priority queue dynamic wavelength and bandwidth allocation algorithm is presented. This algorithm allocates the network resources efficiently and fairly using several techniques, in order to meet the QoS demands of Tactile and other types of healthcare applications. Extensive simulation results indicate the effectiveness of the proposed algorithm to provide the required QoS for medical applications under various simulation scenarios whereas other well-known schemes are shown to lack such support.

The flexible nature of elastic optical networks (EONs) effectively uses spectral resources for optical communication by allocating the minimum required bandwidth to network connections. Since the energy consumption of such networks scales with the magnitude of bandwidth demand, addressing the issue of energy wastage is important. This fact has a profound impact on the design of efficient schemes for energy aware optical networks, and adaptivity arises as one of the most important properties of these networks. Learning Automata are Artificial Intelligence tools that have been used in networking algorithms, when adaptivity to the characteristics of the network environment can result in significantly improved network performance. In this work, a new adaptive power-aware algorithm is introduced, which selectively switches off bandwidth-variable optical transponders (BVTs) under low utilization conditions, to achieve energy efficiency. A novel adaptive scheme, which makes use of Learning Automata to significantly reduce the total energy consumption, while at the same time avoiding the onset of congestion, is proposed. The proposed scheme monitors network congestion, in terms of Bandwidth Blocking Probability (BBP), and the learning mechanism finds the optimal amount of energy-saving so that congestion is avoided, while at the same time significant energy savings are achieved. The proposed Learning Energy-Saving Algorithm (LESA) is evaluated via extensive simulation results, which indicate that it achieves an energy saving of up to 50%, compared to other energy efficient solutions.

Mobile cloud computing is an emerging commercial infrastructure paradigm to enhance the power of smart mobile devices. However, there are significant problems pertaining to efficient provisioning and delivery of mobile applications using cloud IT resources. These barriers concern various levels such as enabling individual mobile user's to dynamically adjust their computing demands and managing these infrequent demands under intermittent network connectivity along with maintaining quality of service (QoS), whereas cloud resources are available for long-time quanta granularity and incur significant setup times. This paper proposes a demand-based computation offloading framework (DBCOF) to balance these competing goals. It provides a mechanism to solve the key problem of application partitioning according to user's demand. In addition, it also reduces the monetary costs per request for the service providers and minimizes service-level agreement (SLA) violations for maintaining QoS. Previous approaches to these problems have focused only on maximizing energy savings in all contexts, irrespective of user's computing requirements. We have implemented the DBCOF prototype for Android and explored its design space by extending CloudSim, the state-of-the-art cloud simulator. Analyses of the results show the effectiveness of DBCOF system in terms of managing computation offloading demands. It also reduces the cost of providing cloud resources to mobile devices to 44.79% and minimizes SLA violations to nearly 1.34% for maintaining the QoS.