Computational fluid dynamics (CFD) simulations play an ever-increasing role in the design and optimization of turbomachinery components. The CFD code TRACE (Turbomachinery Research Aerodynamic Computational Environment) is developed by the German Aerospace Center (DLR) and is DLR’s standard solver for internal flows as well as in industry use by MTU Aero Engines and Siemens Energy. Up to this point TRACE was only partly analyzed using the profiling tool GPTL. In this paper the first efforts in profiling, tracing and analyzing TRACE using the tools Score-P, Vampir, and Cube as well as the subsequent improvements are shown. The analysis moves from the easy pickings to a deeper look at the communication patterns.