Desktop pause every few seconds on Linux - OpenSUSE Leap 15.0

I recently installed the CUDA 10.0 package on OpenSUSE Leap 15.0 following the instructions posted on the nvidia documentation site (my GPU is the Quadro K2200):

Now my desktop GUI pauses for about a second every few (5?) seconds.

During the CUDA installation I noticed that a number of nvidia driver packages were switched from the normal nvidia repository to the CUDA one, namely the following four. Note, the ‘i+’ means that it’s installed on my system, and ‘v’ means a different version of the same package is installed.

$ zypper search -s nvidia | grep '^[iv]
i+ | nvidia-computeG04             | package | 410.48-0                            | x86_64 | cuda                   
v  | nvidia-computeG04             | package | 390.87-lp150.10.1                   | x86_64 | nVidia Graphics Drivers
i+ | nvidia-gfxG04-kmp-default     | package | 410.48_k4.12.14_lp150.12.4-0        | x86_64 | cuda                   
v  | nvidia-gfxG04-kmp-default     | package | 390.87_k4.12.14_lp150.11-lp150.10.1 | x86_64 | nVidia Graphics Drivers
i+ | nvidia-glG04                  | package | 410.48-0                            | x86_64 | cuda                   
v  | nvidia-glG04                  | package | 390.87-lp150.10.1                   | x86_64 | nVidia Graphics Drivers
i+ | x11-video-nvidiaG04           | package | 410.48-0                            | x86_64 | cuda                   
v  | x11-video-nvidiaG04           | package | 390.87-lp150.10.1                   | x86_64 | nVidia Graphics Drivers

Here is the information regarding the nvidia repositories themselves:

$ zypper lr -u | grep nvidia
 1 | cuda                                | cuda                                    | Yes     | (r ) Yes  | Yes     | http://developer.download.nvidia.com/compute/cuda/repos/opensuse15/x86_64      
 2 | download.nvidia.com-leap            | nVidia Graphics Drivers                 | Yes     | (r ) Yes  | Yes     | https://download.nvidia.com/opensuse/leap/15.0

And finally here are the specifics of the kernel installed on the system:

$ uname -r
4.12.14-lp150.12.16-default

Any help would be greatly appreciated

you shouldn’t mix driver components from two different sources. If you want to use CUDA, you should only use driver components from NVIDIA archives/repos.

Follow the instructions in the linux install guide to completely remove all old driver components, and use the proper runfile or package manager install methods for driver + CUDA toolkit, using only NVIDIA sources.

Thanks for the reply. I removed all Nvidia drivers before installing CUDA, and then the reverse when switching back - I had either all Nvidia drivers installed or all CUDA drivers installed.