Energy-Efficient Recovery Algorithm in the Fault-Tolerant Tree-Based Fog Computing (FTBFC) Model

Abstract

In order to reduce the traffic of networks and servers, the fog computing model is proposed to realize the IoT (Internet of Things). In our previous studies, the fault-tolerant tree-based fog computing (FTBFC) model is proposed to be tolerant of faults of fog nodes. If a fog node $f_{j}$ is faulty, child fog nodes of $f_{j}$ are disconnected. Another operational fog node $f_{i}$ at the same level as $f_{j}$ is a candidate parent node of disconnected nodes. One candidate node is selected to be a parent of every disconnected node. However, the new parent node $f_{i}$ has to additionally process data from the disconnected nodes. In order to reduce the energy consumption of each new parent node, we propose a modified FTBFC (MFTBFC) model where disconnected nodes are partitioned into groups and fog nodes in each group are connected to a different candidate node. We also propose an SMPR (selecting multiple parents for recovery) algorithm to select a candidate parent node for each disconnected node so that the electric energy consumption of each new parent node can be reduced. In the evaluation, we show the energy consumption and execution time of each new parent fog node can be reduced in the SMPR algorithm.