A smart grid (SG) is a sophisticatedly integrated hybrid power generating system which allows bidirectional energy as well as management data exchanges. In this paper we look at improving energy efficiency in large scale cooperating power consuming as well as power generating systems. We discuss a clustered as well as hierarchical power scheduling algorithms that are geared towards optimizing the management of power tariffs, storage and distribution in a cooperative environment. From a generation perspective, solar intensity prediction is proposed for power generation forecasting and whereas from a power consumption perspective, we evaluate and model the power consumed by these distributed systems (consumers) and propose improving resource allocation, scheduling and network traffic management so as to make network and computing resources more power efficient.
