Abstract Research into software architecture and design has become more and more prominent since the 1990s. Since then, companies have reported how software architecting helped them to tackle various challenges in system design, especially related to system-level quality properties such as scalability and maintainability. Academic research in software architecture has focused on several areas, including architecture description through views and architecture description languages, and on methods for evaluating architectural designs. While much of the contribution of research in software architecture was inspired by industrial experiences, little of the research was validated beyond individual case studies. Many scientific disciplines are currently harvesting fruits from large-scale data collection about their subjects of study. Therefore, this chapter contributes a discussion of challenges and directions for Big Data-driven studies of software architecture. Given the large amount of effort that is needed for this type of research, a promising direction is to look into a community-based infrastructure for enabling and supporting this type of research. We share lessons learned through building various tools that could form building blocks in such an infrastructure. Based on these, we synthesize a reference architecture for creating such a community-wide infrastructure for Big Data-based research in software architecture.