🚀`ZJUER`: 3D Heat Transfer

3D heat transfer simulation with CUDA.

Result, reference, analysis and more details are all in HeatTrans/README.md

Environment

If you want to launch the demo in this repository, you need to install the following software and equips with a NVIDIA GPU.

Ubuntu 20.04
OpenGL 4.6, GLUT
NVIDIA Driver (GeForce MX150 or Newer, CUDA 11.4)
CMake 3.16.3

Configure

cd 3D-Heat-Transfer/HeatTrans
mkdir build
cmake -DCMAKE_BUILD_TYPE=Release -B build -S .
cmake --build build -j8

Operation

Please run the executable file heatTransfer3D using command below. You can rotate the bunny modle with $\leftarrow$ and $\rightarrow$, zoom in and zoom out with $\uparrow$ and $\downarrow$. Press esc to exit, s to propagete one frame, i to propagte contiously.

./build/heatTransfer3D -f objs/bunny.obj

CUDA Platform

./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "NVIDIA GeForce MX150"
  CUDA Driver Version / Runtime Version          11.4 / 11.4
  CUDA Capability Major/Minor version number:    6.1
  Total amount of global memory:                 4042 MBytes (4238737408 bytes)
  (003) Multiprocessors, (128) CUDA Cores/MP:    384 CUDA Cores
  GPU Max Clock rate:                            1532 MHz (1.53 GHz)
  Memory Clock rate:                             3004 Mhz
  Memory Bus Width:                              64-bit
  L2 Cache Size:                                 524288 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        98304 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Supports MultiDevice Co-op Kernel Launch:      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 11.4, CUDA Runtime Version = 11.4, NumDevs = 1
Result = PASS

X11 Forwarding Service on GPU server

Both on server and client.

Install essential software and xclock testing app.

sudo apt-get install xorg
sudo apt-get install xauth
sudo apt-get install openbox
sudo apt-get install xserver-xorg-legacy
sudo apt install x11-apps # xclock

Please install Remote X11 extension in VScode if you want to use X11 Forwarding in your vscode terminal.

Configure X11 Forwarding.

sudo cp /etc/ssh/sshd_config /etc/ssh/sshd_config.backup
sudo vim /etc/ssh/sshd_config

sshd_config

# sshd_config
...
X11Forwarding yes
ForwardX11 yes
ForwardX11Trusted yes
ForwardAgent yes # or AllowAgentForwarding
...

# restart ssh
systemctl restart sshd
# test X11 forwarding
xclock

Problem Record

I want to use X11 server forwarding function to display result returned from the rented GPU server, but it will occupy the GPU resource. Terminal on GPU server will return Segmentation fault (core dumped). On your personal computer, if you run demos in cuda_by_example/chapter08 provided by NVIDIA, terminal may return:

all CUDA-capable devices are busy or unavailable in ../common/gpu_anim.h at line 77

X11 server cannot provide service simultaneously when running GPU program on my own computer. If you want to run program, please do as follows:

nvidia-settings # type in terminal
PRIME Profiles->NVIDIA(Performance Mode) # Find in NVIDIA X Server Settings GUI

LICENSE

Apache License Version 2.0

HangX-Ma/CUDA-HeatTransfer-3D

🚀ZJUER: 3D Heat Transfer