Your web browser doesn't support some required capabilities.
This interactive simulation works best with the latest version of Chrome, Firefox, or Safari.
An error occurred. Please visit Hands-on Labs Support
to get help or report this issue.Unable to initialize the simulation player:
Please visit Hands-on Labs Support
to get help or report this issue.Drive it with your mouse, your finger, or just use the arrow keys.
Use Learn mode to learn the demo. The orange boxes show where to click.
Use Present mode to hide the orange boxes and notes.
Click a Shortcut to jump to a specific part of the demo.
This part of the lab is presented as a Hands-on Labs Interactive Simulation. This will allow you to experience steps which are too time-consuming or resource intensive to do live in the lab environment. In this simulation, you can use the software interface as if you are interacting with a live environment.
The orange boxes show where to click, and the left and right arrow keys can also be used to move through the simulation in either direction.
In this lab there will be a lot of typing in a putty window. Each command that needs to be type will be highlighted in blue.
We will be installing the NVIDIA vGPU driver (NVIDIA-VMware_ESXi_6.5_Host_Driver 384.43-1OEM.650.0.0.4598673) on the vSphere host (v1-vdi-3d-perf18). To do this, you will need to open a putty session to that host.
For this module, we have already opened a putty session for you.
Note: if the text stops, hit the tab key for command autocompletion
Type cat install_nvidia_vib and press Enter
Type ./install_nvidia_vib `pwd`/NVIDIA-VMware_ESXi_6.5_Host_Driver 384.43-1OEM.650.0.0.4598673.x86_64.vib and press Enter
We can verify the installation of the NVIDIA driver by using the command nvidia-smi
Type nvidia-smi and press Enter
Please take a moment to look through the information about this driver
By default the graphics setting is set to Shared, but for this lab we have selected Shared Direct
We need to add the NVIDIA vGPU as a shared PCI device to a VM ml-vgpu-hol
Lets power on the VM and install NVIDIA linux grid driver
Note: if the text stops, hit the tab key for command autocompletion
Type cd Downloads and press Enter
Type sudo ./NVIDIA-Linux-x86_64-384.43-grid.run and press Enter
We can verify the installation of the NVIDIA driver by using the command nvidia-smi
Note: if the text stops, hit the tab key for command autocompletion
Type nvidia-smi and press Enter
Please take a moment to look through the information about this driver. You will see very similar information here as you did when we ran this command on the host.
Note: if the text stops, hit the tab key for command autocompletion
Type cat install_cuda and press Enter
Type sudo ./install_cuda and press Enter
We will now run deviceQuery to verify the installation of the CUDA driver
Note: if the text stops, hit the tab key for command autocompletion
Type cat ./run_deviceQuery and press Enter
Type ./run_deviceQuery and press Enter
We will now install GPU version 1.2 of TensorFlow.
Note: if the text stops, hit the tab key for command autocompletion
Type cat install_tf and press Enter
Type sudo ./install_tf and press Enter
We can now run the Handwriting Recognition benchmark using TensorFlow
Type cd /home/ml/mnist_runs and press Enter
Type ./run_mnist_one_iter >& log& and press Enter
Verify that the GPU is being used and press Enter
Type nvidia-smi and press Enter
Type tail -f log and press Enter
When the program ends, it prints out the test error percentage. You can see that error rate in recognizing handwritten digits is 0.8%!
To return to the lab, click the link in the top right corner or close this browser tab.