问题
Iam using nvidia gt 440 gpu. It is used for both display and computational purpose which leads to less performance while computation. can i enable it only for computational purpose? if so how can i disable it from using display.
回答1:
It depends -- are you working on Windows or Linux? Do you have any other display adapters (graphics cards) in the machine?
If you're on Linux, you can run without the X Windows Server (i.e., from a terminal) and SSH into the box (or attach your display to another adapter).
If you're on Windows, you need to have a second display adapter. As long as your display is connected to your GeForce 440 GT, there's no way to use it only for computational purposes. That also includes Remote Desktop, which won't work at all unless you have a Tesla card because of the way the WDDM (Windows Display Driver Model) was designed (it can't be accessed from within Session 0, which is where the RDP service runs).
回答2:
I'm using Intel integrated graphics for display purposes and GPU for compute purpose on Linux. You'll need to setup from bios to use the integrated graphics on mobo. This will leave your GPU free. It depends on your hardware available. =)
How much does it affects the performance? I did checked before, the display in windows did takes up some memory (less than 10mb).
回答3:
Check that you have write permission on the /dev/nvidia* devices. The CUDA C Getting Started Guide for Linux contains a script that automatically sets the correct permissions at startup.
来源：https://stackoverflow.com/questions/12455711/how-to-enable-cuda-only-for-computing-purpose-not-for-display