Understanding distributed systems is a complex task. There are many subsystems involved, such as network equipment, disk and CPU, which effect behavior. In order to analyze this kind of applications, different approaches have been proposed: simulation, emulation and experimentation. Each paradigm has evolved independently, providing their own set of tools and methodologies. This paper explores how these tools and methodologies can be combined in practice. Given a simple question on a particular system, we explore how different experimental frameworks can be combined in practice. We use a representative framework for each methodology: Simgrid for simulation, Distem for emulation and Grid'5000 for experimentation. Our experiments are formally described using the workflow logic provided by the XP Flow tool. Our long term goal is to foster a coherent methodological framework for the study of distributed systems. The contributions of this article to that end are the following: we identify a set of pitfalls in each paradigm that experimenters may encounter regarding models, platform descriptions and others. We propose a set of general guidelines to avoid these pitfalls. We show these guidelines may lead to accurate simulation results. Finally, we provide some insight to framework developers in order to improve the tools and thus facilitate this convergence.