70
votes
GPU usage monitoring (CUDA)
You can use nvtop, it's similar to htop but for NVIDIA GPUs. Link: https://github.com/Syllo/nvtop
Install on Ubuntu with sudo apt install nvtop
- 801
41
votes
GPU usage monitoring (CUDA)
Recently, I have written a monitoring tool called nvitop, the interactive NVIDIA-GPU process viewer.
It is written in pure Python and is easy to install.
Install from PyPI:
pip3 install --upgrade ...
- 511
24
votes
Accepted
How can I disable (and later re-enable) one of my NVIDIA GPUs?
Disabling:
The following disables a GPU, making it invisible, so that it's not on the list of CUDA devices you can find (and it doesn't even take up a device index)
nvidia-smi -i 0000:xx:00.0 -pm 0
...
- 11k
16
votes
How to automatically force full composition pipeline for Nvidia GPU driver?
My laptop is equipped with a Quadro K2100M, running Ubuntu Bionic with KDE Plasma 5.17 (from the neon repo) and using the nvidia-driver-430.
Since I have a different set of configurations for the ...
- 161
15
votes
GPU usage per process on a Linux machine (CUDA)
I stumbled upon the following:
watch -n 0.5 nvidia-smi pmon -c 1
Though it seems to have a different measure of GPU usage than the volatility metric in the usual nvidia-smi.
Full options in the ...
- 423
14
votes
Accepted
What's the difference between DRM and a graphics driver?
"Graphics driver" can mean any number of things.
The way X (the graphical windowing system) works is that there is a central X server, which can load modules ("X drivers") for different hardware. ...
- 33.4k
11
votes
Accepted
How to disable Hardware Acceleration in Linux?
You can configure Xorg to disable OpenGL / GLX.
For a first try, you can run a second X session: switch to tty2, log in and type:
startx -- :2 vt2 -extension GLX
To permanently disable hardware ...
- 2,575
9
votes
Accepted
Is it possible to completely turn off nvidia GPU to be able to run wayland?
The answer was simple: just install nvidia drivers, open the nvidia settings page and set to use the Intel HD GPU only. Login again and you are done. Works perfectly. Battery lasts much much longer ...
- 459
9
votes
Accepted
How can I tell what version of OpenGL my machine supports on Arch Linux?
$ sudo pacman -S mesa-demos
$ glxinfo | grep "OpenGL version"
- 106
9
votes
Reduce AMD GPU Wattage
Introduction
To reduce power usage you will need to "undervolt", reducing the voltage of your GPU or limit the power states of the GPU.
Using a modern (4.17 or greater) kernel and the ...
- 5,430
8
votes
Accepted
How do I interpret the output of intel-gpu-top and intel-gpu-overlay?
Taken from the link given in the comments in OP.
I was curious as well, so here are just a few things I could grab from the reference manuals. Also of interest is the intel-gpu-tools source, and ...
- 1,927
8
votes
Accepted
What are the advantages of hardware-accelerated terminal emulators?
They can potentially be faster at outputting and refreshing vast amounts of information. It could also allow for smooth(er) scrolling. Human beings however are quite slow at reading this information, ...
- 32.9k
8
votes
How can I test my GPU RAM integrity?
Dicslosure: I'm an author of the program linked below, so my opinion can be weighted
memtest_vulkan is small & simple open-source tool for stress-testing gpu memory on any vulkan 1.1 compatible ...
- 261
7
votes
Accepted
How to check an application is using hardware or software based display rendering?
There are a number of issues here.
First, applications can support a variety of rendering methods, e.g. OpenGL and software rendering. If an application renders in software, then no hardware ...
- 481k
7
votes
Accepted
lm-sensors meaning of output
“hyst” stands for hysteresis. In a sensor’s configuration, it’s the threshold below which the temperature must return for a sensor to no longer be considered critical. In your case, if the “edge” ...
- 481k
6
votes
GPU usage monitoring (CUDA)
I have had processes terminate (probably killed or crashed) and continue to use resources, but were not listed in nvidia-smi. Usually these processes were just taking gpu memory.
If you think you ...
- 163
6
votes
System Monitor in Ubuntu 18 shows only 5.8 GB RAM out of 8 GB installed
Quoting this Quora article:
Installed RAM is the actual physical memory installed in your PC.
5.88 usable = Let’s just round this to 6GB for simplicity
8GB - 6GB = 2GB. This means that 2GB is being ...
- 31.1k
6
votes
Accepted
How to shift process from GPU to CPU usage
Your GPU is being used for both display and compute processes; you can see which is which by looking at the “Type” column — “G” means that the process is a graphics process (using the GPU for its ...
- 481k
5
votes
How to adjust NVIDIA GPU fan speed on a headless node?
I've written a pip-installable Python script to do something similar to @AlexsandrDubinsky's suggestion.
When you run fans.py, it sets up a temporary X server for each GPU with a fake display ...
- 151
5
votes
GPU usage monitoring (CUDA)
Glances has a plugin which shows GPU utilization and memory usage.
http://glances.readthedocs.io/en/stable/aoa/gpu.html
Uses the nvidia-ml-py3 library: https://pypi.python.org/pypi/nvidia-ml-py3
5
votes
Accepted
How to prevent GPU from overheating and auto turning off
I found the solution, there are some files on /sys/class/drm/card0/device the file pp_dpm_mclk indicates GPU memory clock, and the file pp_dpm_sclk indicates GPU core clock, mine:
$ egrep -H . /sys/...
- 646
5
votes
Accepted
rx 560x slower than integrated vega gpu on fedora 28
I regard the same issue as you on the Arch Linux (mesa-git, llvm-svn, linux 4.18.12 (even running 4.19.x or mainline 4.20rcx)). Stable linux kernel isn't, I would say, optimized for our gpu yet. ...
- 66
5
votes
How to monitor AMD GPU on Linux Server
As a simple user you can acces to /sys/class/drm/ which gives you many informations about all the gpus in your system.
They are listed as card0 card1 and so on..
I'm using Radeon VII as a simple ...
- 231
5
votes
Reduce AMD GPU Wattage
I'm going to complement kemotep's answer, in connection with my observations.
A small preface: English is not my native language. Please ignore semantic, syntactical errors and incorrectly chosen ...
- 51
5
votes
Accepted
An equivalent of Looking Glass where VM side runs Linux?
I am the Author of Looking Glass.
The project already has Linux guest support as the host application is agnostic and can be built for both. Please note though that the support for the Linux guest is ...
- 351
4
votes
Accepted
What is the difference between optirun and primusrun (bumblebee)
Bumblebee, earlier, was using VirtualGL as its core and now switched to Primus technology.
optirun uses VirtualGL and primusrun uses the Primus technology. That must be the reason.
Note: Though the ...
- 525
4
votes
GPU usage monitoring (CUDA)
To monitor GPU usage in real-time, you can use the nvidia-smi command with the --loop option on systems with NVIDIA GPUs. Open a terminal and run the following command:
nvidia-smi --query-gpu=...
- 41
4
votes
How do I check what drivers / layers of code are being used for my GPU?
I've seen this question asked a lot but it seems to always be answered wrong. You can simply look at the documentation of your respective distribution and find out which based on package names. But ...
- 5,257
4
votes
Ethernet not working after removing GPU
When you removed the GPU, the PCIe bus configuration of your system changed, causing the bus location of your NIC to change. This change is reflected in the PCI IDs in the lspci listing, and also in ...
- 114k
Only top scored, non community-wiki answers of a minimum length are eligible