A Dynamic Tracing Mechanism for Performance Analysis of OpenMP Applications

Authors:
Jordi Caubet;Judit Gimenez;Jesús Labarta;Luiz De Rose;Jeffrey S. Vetter
Affiliations:
-;-;-;-;-
Venue:
WOMPAT '01 Proceedings of the International Workshop on OpenMP Applications and Tools: OpenMP Shared Memory Parallel Programming
Year:
2001

Citing 3
Cited 6

The Paradyn Parallel Performance Measurement Tool

Computer
The Dynamic Probe Class Library: An Infrastucture for Developing Instrumentation for Performance Tools

IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium
An API for Runtime Code Patching

International Journal of High Performance Computing Applications

VGV: Supporting Performance Analysis of Object-Oriented Mixed MPI/OpenMP Parallel Applications

IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
CATCH - A Call-Graph Based Automatic Tool for Capture of Hardware Performance Metrics for MPI and OpenMP Applications

Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
On the Scalability of Tracing Mechanisms

Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
Scalable fine-grained call path tracing

Proceedings of the international conference on Supercomputing
A performance measurement infrastructure for co-array fortran

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing
Runtime function instrumentation with EZTrace

Euro-Par'12 Proceedings of the 18th international conference on Parallel processing workshops

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present OMPtrace, a dynamic tracing mechanism that combines traditional tracing with dynamic instrumentation and access to hardware performance counters to create a powerful tool for performance analysis and optimization of OpenMP applications. Performance data collected with OMPtrace is used as input to the Paraver visualization tool for detailed analysis of the parallel behavior of the application. We demonstrate the usefulness of OMPtrace and the power of Paraver for tuning OpenMP applications with a case study running the US DOE ASCI Sweep3D benchmark on the IBM SP system at the Lawrence Livermore National Laboratory.