In this paper we address the problem of reducing the energy consumption in distributed embedded systems associated with time-constraints and equipped with fault-tolerant techniques. A greedy heuristic is presented to reduce the energy during task mapping and fault tolerance policy assignment. Fault tolerance is achieved through task re-execution and replication. The proposed approach can obtain much lower energy consumption compared to the solution without considering energy while tolerating the same number of faults and satisfying the time-constraints. The effectiveness of the proposed approach is evaluated by using extensive experiments.