TY - JOUR
T1 - Performance analysis of GYRO
T2 - A tool evaluation
AU - Worley, P.
AU - Candy, J.
AU - Carrington, L.
AU - Huck, K.
AU - Kaiser, T.
AU - Mahinthakumar, G.
AU - Malony, A.
AU - Moore, S.
AU - Reed, D.
AU - Roth, P.
AU - Shan, H.
AU - Shende, S.
AU - Snavely, A.
AU - Sreepathi, S.
AU - Wolf, F.
AU - Zhang, Y.
PY - 2005/1/1
Y1 - 2005/1/1
N2 - 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 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.
AB - 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 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.
UR - http://www.scopus.com/inward/record.url?scp=24344475222&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/16/1/076
DO - 10.1088/1742-6596/16/1/076
M3 - Article
AN - SCOPUS:24344475222
SN - 1742-6588
VL - 16
SP - 551
EP - 555
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - 1
ER -