To tackle the scale issue between the increasing number of IoT devices and the limited number of datacenters, new computing paradigms have emerged, such as edge and fog computing. Numerous and heterogeneous computing approaches and techniques can be deployed to such environments, making their seamless integration and configuration a tedious and error-prone task. In this paper, we argue that tools and methods originating from software product lines engineering can assist stakeholders in tackling this configuration problem. In particular, we show that edge and fog environments can be represented with feature modeling, and thus existing tools can be leveraged to predict and optimize performances of edge and fog environments.