Skip to content

TensorWorks/UE4-Container-Performance-Profiler

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

UE4 Container Performance Profiler

DISCLAIMER

The dockerfiles assume that you will provide a packaged UE4 project in the appropriate directory:
For Linux: Linux/LinuxNoEditor
For Windows: Windows/WindowsNoEditor

TL:DR I Just Wanna Run It

For your convenience, we've provided docker images that will use CSV profiler on Epic Games' Subway Sequencer Demo.
By default, the profiling of GPU stats is enabled with the following metrics being captured:

  • FrameTime
  • RenderThreadTime
  • GameThreadTime
  • GPUTime
  • RHIThreadTime
  • MemoryFreeMB
  • PhysicalUsedMB
  • VirtualUsedMB

Docker Hub

https://hub.docker.com/r/belchy06/sequence_profiler

Windows (21h1 only)

In a terminal, run:
docker run -it --rm --isolation=process --device class/5B45201D-F2F2-4F3B-85BB-30FF1F953599 -v d:/data/profiler/output:c:/output belchy06/sequence_profiler:windows

Linux

In a terminal, run:
docker run --rm --gpus=all -v $HOME/output:/output belchy06/sequence_profiler:linux

Running Multiple Containers With docker-compose

Windows

You can run the profiler through the windows docker-compose file in a powershell terminal with the following command:

$env:SOURCE="c:/Directory/For/Output/Files"; docker-compose up --scale profiler=N

Replace the SOURCE value with the directory in which you want the output files to be located
Repalce the N argument with the number of profilers you want to run simultaneously

Linux

Currently the docker-compose file does not work as I am still searching for the equivalent of the docker run arg -gpus all for the yml file.

About

A profiler for Unreal Engine container performance

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages