CDF computing and event data models

The authors discuss the computing systems, usage patterns and event data models used to analyze Run II data from the CDF-II experiment at the Tevatron collider. A critical analysis of the current implementation and design reveals some of the stronger and weaker elements of the system, which serve as lessons for future experiments. They highlight a need to maintain simplicity for users in the face of an increasingly complex computing environment