TY - GEN
T1 - An open-source tool-chain for performance analysis
AU - Coulomb, Kevin
AU - Degomme, Augustin
AU - Faverge, Mathieu
AU - Trahay, François
PY - 2012/12/1
Y1 - 2012/12/1
N2 - Modern supercomputers with multi-core nodes enhanced by accelerators as well as hybrid programming models introduce more complexity in modern applications. Efficiently exploiting all of the available resources requires a complex performance analysis of applications in order to detect time-consuming or idle sections. This paper presents an open-source tool-chain for analyzing the performance of parallel applications. It is composed of a trace generation framework called EZTRACE, a generic interface for writing traces in multipe formats called GTG, and a trace visualizer called VITE. These tools cover the main steps of performance analysis - from the instrumentation of applications to the trace analysis - and are designed to maximize the compatibility with other performance analysis tools. Thus, these tools support multiple file formats and are not bound to a particular programming model. The evaluation of these tools show that they provide similar performance compared to other analysis tools.
AB - Modern supercomputers with multi-core nodes enhanced by accelerators as well as hybrid programming models introduce more complexity in modern applications. Efficiently exploiting all of the available resources requires a complex performance analysis of applications in order to detect time-consuming or idle sections. This paper presents an open-source tool-chain for analyzing the performance of parallel applications. It is composed of a trace generation framework called EZTRACE, a generic interface for writing traces in multipe formats called GTG, and a trace visualizer called VITE. These tools cover the main steps of performance analysis - from the instrumentation of applications to the trace analysis - and are designed to maximize the compatibility with other performance analysis tools. Thus, these tools support multiple file formats and are not bound to a particular programming model. The evaluation of these tools show that they provide similar performance compared to other analysis tools.
U2 - 10.1007/978-3-642-31476-6_4
DO - 10.1007/978-3-642-31476-6_4
M3 - Conference contribution
AN - SCOPUS:84881409287
SN - 9783642314759
T3 - Proceedings of the 5th International Workshop on Parallel Tools for High Performance Computing 2011
SP - 37
EP - 48
BT - Proceedings of the 5th International Workshop on Parallel Tools for High Performance Computing 2011
T2 - 5th International Workshop on Parallel Tools for High Performance Computing, HPC 2011
Y2 - 26 September 2011 through 27 September 2011
ER -