½-1 day course to help you understand why & where your computer models may not use every available CPU cycle and what you can do about it. We focus on examining serial codes with hands-on examples to help you spot and correct performance issues in a FORTRAN code. For the full day version, we also cover looking at performance issues with parallel codes that use OpenMP or MPI. Profilers covered include gprof/gcov, TAU, Allinea MAP/Forge and Intel Parallel Studio (depending which is available at your institution).

Related Training

Also see our main training page