Tutorial

Analyze MPI Applications with Intel® Trace Analyzer and Collector and Intel® VTune™ Profiler

Download PDF
ID 773180
Date 3/31/2023
Public
Document Table of Contents
Document Table of Contents x
Analyze MPI Applications with Intel® Trace Analyzer and Collector and Intel® VTune™ Profiler
Analyze MPI Applications with Intel® Trace Analyzer and Collector and Intel® VTune™ Profiler x
Prerequisites Starting Intel® Trace Analyzer Starting Intel® VTune™ Profiler Analyzing an MPI Application Summary Notices and Disclaimers
Analyzing an MPI Application x
Optimizing MPI Communications Improving Intra-process Performance
Optimizing MPI Communications x
Prepare for Analysis Ungroup MPI Functions Detect Serialization in Function Profile and Message Profile Compare Original Trace File With Idealized Trace File Remove Serialization Compare Two Trace Files Analyze Optimized Communications
Improving Intra-process Performance x
Run Basic Hotspots Analysis Interpret Results
Analyze MPI Applications with Intel® Trace Analyzer and Collector and Intel® VTune™ Profiler

Remove Serialization

You can improve the performance of the poisson sample program by replacing the blocking  MPI_Sendrecv with non-blocking communications via MPI_Isend. The modified source file pardat.f90_icomm is available in the source folder. The trace file of the modified program is also available: <install-dir>/examples/traces/poisson_icomm.single.stf.

Once corrected, the single iteration of the revised program will look similar to this:



Since poisson_sendrecv.single.stf is a striking example of serialization, almost all of the Intel® Trace Analyzer charts show this interesting pattern. But in the real-world cases, it may be necessary to formulate a hypothesis regarding how the program should behave and to check this hypothesis using the most suitable chart.

Parent topic: Optimizing MPI Communications
Level Two Title