Tutorial: Analyze Common Performance Bottlenecks using Intel VTune Profiler in a C++ Sample Application - Linux* OS

ID 762029
Date 10/15/2021
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Tutorial: Analyze Common Performance Bottlenecks with Intel® VTune™ Profiler - C++ Sample Code (Linux* OS)

Discover how to use Intel® VTune™ Profiler for Linux* OS to identify algorithm or hardware utilization issues that can cause your applications to spend large amounts of time performing tasks and underutilize available hardware resources.

About This Tutorial

This tutorial guides you through the steps required to analyze and optimize a sample matrix application that performs multiplication of large matrices. It introduces you to the main concepts of VTune Profiler and the iterative process of analyzing and optimizing an application.

The tutorial was last updated for the Intel VTune Profiler 2021 product release.

Estimated Duration

20-30 minutes.

Learning Objectives

After you complete this tutorial, you should be able to:

  • Open the pre-configured matrix sample project in VTune Profiler.
  • Run the Performance Snapshot analysis to locate the main problem areas in the matrix sample application and identify next steps for optimization.
  • Run the Hotspots and Memory Access analyses to better understand the main bottleneck and determine next steps.
  • Navigate the source code from inside VTune Profiler to locate the lines of code with memory access bottlenecks.
  • Use the HPC Performance Characterization analysis to identify microarchitecture underutilization issues related to lack of proper vectorization.
  • Compare results before and after optimization.

More Resources

Start Here