Future Mobile Networks (MNs), 5G and beyond 5G, will require a paradigm shift from traditional resource allocation mechanisms as Base Stations (BSs) will be empowered with computation capabilities (i.e., offloading and computation is performed closer to mobile users). This is motivated by the expected data explosion in the volume, variety, and velocity, generated by pervasive mobile and Internet of Things (IoT) devices at the network edge. Towards efficient resource management, within the Multi-access Edge Computing (MEC) paradigm, we make use of the Long Short-Term Memory (LSTM) neural network for time series forecasting and control-theoretic techniques for foresighted optimization, thus bringing intelligent mechanisms for handling network resources within the network edge. Here, we propose online algorithms for autoscaling and reconfiguring the computing-plus-communication resources within the virtualized computing platform, and also enable dynamic switching on/off BSs by taking into account the forecasted traffic load and harvested energy. The main goal is to minimize the overall energy consumption, with a guarantee of Quality of Service (QoS). Our numerical results, obtained through trace-driven simulations, show that the proposed optimization strategies lead to a considerable reduction in the energy consumed by the edge computing and communication facilities, promoting energy self-sustainability within the MN through the use of green energy.