Getting no vga output in Centos 7 when using a Tesla M4 video card

I have a Centos 7.2 system and am using a Tesla M4 video card with the system.
Since the Tesla M4 has no output (no VGA/DVI) I am using the onboard VGA.
The issue is when I install the nvidia driver, the VGA output goes away and if I uninstall the driver VGA is back. I am able to set the IP address, and then install the nvidia driver and use the sample sdk so I know the video card is working.

Do I need to do something to switch the display on or forward the video to the onboard VGA?

| NVIDIA-SMI 352.99 Driver Version: 352.99 |
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| 0 Tesla M4 Off | 0000:02:00.0 Off | 0 |
| N/A 32C P8 7W / 75W | 44MiB / 3839MiB | 0% Default |

| Processes: GPU Memory |
| GPU PID Type Process name Usage |
| 0 3101 G /usr/bin/Xorg 0MiB |
| 0 97297 G /usr/bin/X 12MiB |
| 0 97528 G /usr/bin/gnome-shell 10MiB |


Your best bet is to get the X server to run completely in the background by passing it the -sharevts and -novtswitch options. Note that by default, the X server will still try to listen for and respond to keyboard and mouse events, so you might need to configure it to ignore those.

If you’re just trying to use the GPU for compute, you shouldn’t need to start an X server at all, or at least configure it to use the onboard VGA device and its driver rather than the nvidia one.