Intel® VTune™ Profiler Tutorials
Published: 12/07/2017
Last Updated: 12/11/2019
Current Tutorials
The following tutorials are quick paths to start using the Intel® VTune™ Profiler. Each demonstrates an end-to-end workflow you can ultimately apply to your own applications.
Take This Short Tutorial | Learn To Do This | |
---|---|---|
Finding Common Performance Bottlenecks |
C++ Tutorial |
Understand where your application is spending time, identify the most time-consuming program units (hotspots), and detect hardware usage bottlenecks for a sample application. |
Analyzing an OpenMP* and MPI Application |
C++ Tutorial |
Identify issues in a hybrid OpenMP and MPI application using MPI Performance Snapshot, Intel Trace Analyzer and Collector, and Intel VTune Profiler. |
TIP: For more end-to-end analysis use cases, explore the Intel VTune Profiler Cookbook.
NOTE: Intel® VTune™ Amplifier has been renamed to Intel® VTune™ Profiler starting with its version for Intel® oneAPI Base Toolkit (Beta). You can still use a standalone version of the VTune Profiler, or its versions integrated into Intel Parallel Studio XE or Intel System Studio.
Deprecated Tutorials
The following tutorials are deprecated and have not been updated for the latest Intel VTune Profiler release. The features described may be different in newer versions of Intel VTune Profiler.
Take This Short Tutorial | Learn To Do This | |
---|---|---|
Finding Hotspots |
Fortran Tutorial |
Identify where your application is spending time, detect the most time-consuming program units and how they were called. |
Analyzing Locks and Waits |
C++ Tutorial |
Identify locks and waits preventing parallelization. |
Identifying Hardware Issues |
C++ Tutorial |
Identify the hardware-related issues in your application such as data sharing, cache misses, branch misprediction, and others. |
Analyzing Disk Input/Output Waits |
C++ Tutorial |
Analyze an I/O bound application that uses the system file cache and user buffer to work with the I/O device. |
Identifying False Sharing |
C Tutorial |
Identify false sharing. |
Enabling Performance Collection on an Embedded Linux* System |
C++ Tutorial |
Configure a remote Linux embedded system built with the Yocto Project* 2.1 environment for application analysis with VTune Amplifier sampling drivers. Analyze where your application is spending time and identify the most time-consuming program units with Advanced Hotspots analysis. |
Finding Hotspots on an Android* Platform |
C++ Tutorial |
Configure and run a remote Basic Hotspots analysis on an Android target system. |
Analyzing Energy Usage on an Android* Platform |
Run the Energy analysis with the Intel SoC Watch collector (available with the Intel System Studio) directly on the target Android system and view the collected data with the VTune Amplifier installed on the host Windows* or Linux* system. |
|
Analyzing Energy Usage on a Windows* Platform |
Tutorial |
Run the Energy analysis of an idle system and a sample application with the Intel SoC Watch collector (available with Intel System Studio) directly in the target Windows* system. Copy the results to the Windows host system and view the collected data with VTune Amplifier. |
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.