GPU 状态检测 & Burn

GPU 状态检测 & Burn

目录

Linux下Gpu查看

lspci指令

Nvidia自带工具

基于该工具获取GPU内存使用率、GPU占用率：

现成的软件工具

C++开源救命

检测gpu状态(python开源)

 pytorch

opengl/vulkan

Others

---

Linux下Gpu查看 lspci指令

https://linux.die.net/man/8/lspci

lspci is a utility for displaying information about PCI buses in the system and devices connected to them.

lspci | grep -i vga

使用nvidia GPU可以： lspci | grep -i nvidia

Nvidia自带工具

Nvidia自带一个命令行工具可以查看显存的使用情况：

 NVIDIA System Management Interface | NVIDIA Developer

watch -n 0.1 nvidia-smi   //0.1s更新一次

https://developer.download.nvidia.com/compute/DCGM/docs/nvidia-smi-367.38.pdf

基于该工具获取GPU内存使用率、GPU占用率：

现成的软件工具

UNIGINE Benchmarks

GpuTest - Cross-Platform GPU Stress Test and OpenGL Benchmark for Windows, Linux and OS X | Geeks3D.com

C++开源救命

https://github.com/wilicc/gpu-burn

// runLength : unit is seconds, the time of buring GPU.
// useDoubles: true: use double type as array element type.
// useBytes:  -80: 程序会占用GPU当前的剩余memory的80%；
//             1300：程序会占用GPU的1300M byte.
template void launch(int runLength, bool useDoubles, bool useTensorCores, ssize_t useBytes)

Note: 如果有三个GPU,程序会启动三个进程用于burn 每个GPU, 主进程通过pipe与它们通信；

检测gpu状态(python开源)

https://github.com/wookayin/gpustathttps://github.com/wookayin/gpustat

 
pytorch

https://pytorch.org/ 

opengl/vulkan

Others

Running Python script on GPU. - GeeksforGeekshttps://www.geeksforgeeks.org/running-python-script-on-gpu/

https://medium.com/geekculture/executing-a-python-script-on-gpu-using-cuda-and-numba-in-windows-10-1a1b10c29c9https://medium.com/geekculture/executing-a-python-script-on-gpu-using-cuda-and-numba-in-windows-10-1a1b10c29c9

     设置使用哪个gpu: (好像不work)如何指定程序在某个GPU上运行 python_Diana_Z的博客-CSDN博客_python在gpu上运行https://blog.csdn.net/Diana_Z/article/details/89449186

 Is a GPU available? – Machine Learning on GPUhttps://hsf-training.github.io/hsf-training-ml-gpu-webpage/02-whichgpu/index.html

https://www.cyberciti.biz/open-source/command-line-hacks/linux-gpu-monitoring-and-diagnostic-commands/

https://www.mdpi.com/2079-3197/8/1/4/htm