Energy-efficient Cache Updating Scheme with Popularity Prediction.

Content popularity evolution impacts optimization of cache placement and therefore influences design of caching policies. In this paper, we propose a cache updating scheme with popularity prediction and energy cost consideration. Firstly, we use linear regression method to predict future content popularity. Furthermore, we update the preceding cache placement according to predicted popularity aiming at minimizing energy cost. The problem is formulated as a convex optimization problem. Then we transform cache updating cost function, such that the optimal cache updating solution can be gained by Karush-Kuhn-Tucker (KKT) conditions. At last, the simulation results show that our scheme gets better performance in terms of energy consumption compared to the scheme without cache updating energy cost consideration.