Difference between revisions of "GPU621/VTuners"

From CDOT Wiki
Jump to: navigation, search
(Algorithm Optimization)
(Algorithm Optimization)
Line 33: Line 33:
 
==== Hotspot Analysis ====
 
==== Hotspot Analysis ====
  
[File:Hotspot Analysis.png|600px|frame|Here we can see the summary of the output of the Vtune Profiler which shows some general information about the run time and thread usage]
+
[[File:Hotspot Analysis.png|600px|frame|Here we can see the summary of the output of the Vtune Profiler which shows some general information about the run time and thread usage]]
  
 
== Microarchitecture and Memory Bottlenecks ==
 
== Microarchitecture and Memory Bottlenecks ==

Revision as of 15:51, 27 November 2022

Intel Vtune Profiler

Group Members

  1. Mitchell Gregoris
  2. Roger Wu
  3. Nathaniel Ko

Vtune Profiler Features

The Vtune Profiler has a variety of features that provide information to assist in the optimization of application performance, system performance. The profiler also assists in system configuration for HPC, Cloud, IoT, media, storage, etc.

The profiler provides compatibility for a variety of systems and platforms that include the following:

    CPU, GPU, and FGPA
    Any combination of the following languages: SYCL, C, C++, C+, Fortran, OpenCL, Python, Google Go, Java, .NET, Assembly
    Optimized performance that avoids power or thermal throttling 
    Collection of coarse-grained data over extended periods with details results including mapping to source code

Algorithm Optimization

Analyzing Hot Code Paths

Flame Graphs

Here we can see an example of a flame graph selecting the “hottest” processes run in an application which represent 93.3% of the total CPU usage time


Analyzing Hot Spots

Hotspot Analysis

Here we can see the summary of the output of the Vtune Profiler which shows some general information about the run time and thread usage

Microarchitecture and Memory Bottlenecks

Accelerators and XPUs

Parallelism

Platform and I/O

Multi-Node

Vtune Profiler Coding Excercise