P22: Tools or No tools?: A parallel performance analysis case study Session: Posters: Performance Event Type: Poster Time: 5:00pm - 7:00pm Speaker(s): Patrick Worley, Jeff Candy, Laura Carrington, Kevin Huck, Tim Kaiser, Kumar Mahinthakumar, Allen Malony, Dan Reed, Philip Roth, Hongzhang Shan, Sameer Shende, Allan Snavely, Sarat Sreepathi, Ying Yang Location: 4th Floor South Lobby Abstract: The performance of a popular Eulerian gyrokinetic-Maxwell solver code, GYRO, is analyzed on five high performance computing systems. First, a manual approach is taken, using custom scripts to analyze the output of embedded wallclock timers, floating point operation counts collected using hardware performance counters, and traces of user and communication events collected using the profiling interface to Message Passing Interface (MPI) libraries. Parts of the analysis are then repeated or extended using a number of sophisticated performance analysis tools: IPM, KOJAK, SvPablo, TAU, and the PMaC modeling tool suite. The paper briefly discusses what has been discovered via this manual analysis process, what performance analyses are inconvenient or infeasible to attempt manually, and to what extent the tools show promise in accelerating or significantly extending the manual performance analyses. Chair/Speaker Details: Patrick Worley Oak Ridge National Laboratory Jeff Candy General Atomics Laura Carrington San Diego Supercomputer Center Kevin Huck University of Oregon Tim Kaiser San Diego Supercomputer Center Kumar Mahinthakumar North Carolina State University Allen Malony University of Oregon Dan Reed University of North Carolina at Chapel Hill Philip Roth Oak Ridge National Laboratory Hongzhang Shan Lawrence Berkeley National Laboratory Sameer Shende University of Oregon Allan Snavely San Diego Supercomputer Center Sarat Sreepathi North Carolina State University Ying Yang University of North Carolina at Chapel Hill