The above figure shows a Vampir display of an application written using HPF for data parallelism and Opus for task parallelism. The HPF compiler produces Fortran 90 data paralllel modules which execute on multiple processes. The processes interoperate using the Opus runtime system which is built on MPI and pthreads. In systems of this type, it is important to be able to see the influence of different software levels. TAU is able to capture performance data at different parts of the Opus/HPF system exposing the bottlenecks within and between levels.