Measurement-based Research, Sustainability and Research Data Management

This presentation was given at the 6G Conference in Berlin on July 2, 2025, during the session on large-scale experimentation facilities. The presentation showed the benefits of large-scale experimental facilities in research and of shared infrastructures as an enabler of effective research, including specialized hardware setups. This is particularly important in the context of power consumption and sustainability measurements. The presentation demonstrates that realism at scale, stress testing, scalability, and diversity of environments are important factors since many networking phenomena only appear in large, heterogeneous topologies. Consequently, large-scale experimental facilities will advance networking research. At the same time, they will allow reproducibility across sites. Research data management frameworks like NFDIxCS complement the facilities by allowing the organization, storage, preservation, and sharing of research data, particularly research software, in execution environments through the FAIR principle (findable, accessible, interoperable, and reusable). The talk shows that researchers need to address several issues to conduct controlled, repeatable experiments. A large-scale shared experimental facility, combined with research data management, is essential for reproducibility, sustainability, and the long-term value of research, enabling novel experiments in distributed systems and driving scientific and societal progress.