An Algorithm to Change the TBFC Model to Reduce the Energy Consumption of the IoT

The TBFC (Tree-Based Fog Computing) model is proposed to energy-efficiently realize the FC model. Here, fog nodes are tree-structured and support application processes for sensor data. A target node is a most congested one in a TBFC tree. In our approach to reducing the total energy consumption of the IoT, a TBFC tree is changed by applying am operation op, i.e. migrate, replicate, or expand on a target node f. Processes of the node f migrate to child nodes. The node f and each child node are host and guest nodes of the processes, respectively. In addition, the node f is replicated to multiple sibling replicas and is expanded, i.e. new child nodes are created. In addition to the target node f, newly created nodes and guest nodes of processes migrating are affected by applying an operation since the energy to be consumed by the affected nodes also changes. First, we newly propose formulas to estimate the total energy consumption of all the affected nodes. Then, an algorithm is proposed to decide on an operation op on a target node f by using the estimation formulas. In the evaluation, we show the energy consumption of a CC (Cloud Computing) model can be reduced to 0.3 to 0.5% by changing to a TBFC tree of fog nodes in the proposed algorithm.