Use this library of tutorials and guided code samples to improve your understanding of Intel® VTune™ Profiler. Follow procedures described in these materials to run analyses and interpret results.
Linux* version: HTML
Windows* version: HTML
Sample: Pre-built matrix C++ matrix multiplication application. This application is available from the Project Navigator when you first launch Intel VTune Profiler. You can access the sample code from:
Linux: <install-dir>/samples/en/C++/matrix
Windows: <install-dir>\samples\en\C++\matrix
Learning Objectives:
Demonstration: Iterative optimization with VTune Profiler to find bottlenecks in algorithmic and hardware utilization
Performance issues: Memory access, Vectorization
Analysis Types: Performance Snapshot, Hotspots, Memory Access, HPC Performance Characterization, Microarchitecture Exploration
Linux* version: HTML
Sample: heart_demo C++ application that simulates electrophysiological heart activity. You can access the sample code at https://github.com/AlexeyMalkhanov/Cardiac_demo.
Learning Objectives:
Demonstration: Identify issues in a hybrid OpenMP and MPI application.
Tools used: Application Performance Snapshot (APS), Intel Trace Analyzer and Collector
Analysis Types: HPC Performance Characterization analysis
Find guided samples for VTune Profiler in the code sample library on Github.
The VTune Profiler Performance Analysis Cookbook features a collection of technical recipes that target typical use cases and common challenges in software performance analysis.
Copy the archive file from the installation directory to a writable directory or share on your system.
Extract the sample from the archive.
Samples are non-deterministic. Your screens may vary from the screen shots shown throughout these tutorials.
Samples are designed only to illustrate the VTune Profiler features and do not represent best practices for tuning any particular code. Results may vary depending on the nature of the analysis and the code to which it is applied.