Performance Analysis of GYRO: A Tool Evaluation
The performance of the 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 wall clock 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.
- Research Organization:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Director. Office of Science. Office of AdvancedScientific Computing Research. Mathematical Information and ComputationalSciences, Contracts DE-AC02-05CH11231 and DE-FG03-95ER54309 andDE-FC02-04ER25612 and DE-FC02-01ER25491 and DE-FG02-05ER23680 andDE-FG03-01ER25501 and DE-FC02-01ER25490 and DE-AC05-00OR22725
- DOE Contract Number:
- DE-AC02-05CH11231
- OSTI ID:
- 862126
- Report Number(s):
- LBNL-58413; R&D Project: KS3110; BnR: KJ0101030; TRN: US200602%%68
- Resource Relation:
- Conference: SciDAC Scientific Discovery through AdvancedComputing, 2005, San Francisco, CA, June 26-30,2005
- Country of Publication:
- United States
- Language:
- English
Similar Records
A verification of the gyrokinetic microstability codes GEM, GYRO, and GS2
TAU Gateway: Performance Engineering for Science Gateways