All Deep Learning tools are disabled if no suitable graphics card is found. Otherwise the GPU should theoretically always be used. The current Dragonfly version (2020.1) only supports single GPU training and inference. If more than one card is present, the one used for deep learning can be selected from the Dragonfly Preferences:
If you still suspect that the GPU is not used, you may use the following two utilities to verify.
We recommend utilizing Nvidia's SMI utility since it is available on both Windows and Linux and should be installed with the Nvidia driver. On a Windows system it is usually located at
C:\Program Files\NVIDIA Corporation\NVSMI\nvidia-smi.exe
Open a windows command prompt and navigate to that directory. Then run
nvidia-smi.exe -l 5
On a Linux system open a command prompt run
nvidia-smi -l 5
This will show you a screen like this one, that updates every five seconds:
Here we can see various information about the state of the GPUs and what they are doing. Of specific interest in this case is the "Pwr: Usage/Cap" and "Volatile GPU-Util" columns. If your model is indeed using the/a GPU these columns should increase "instantaneously" once you start training the model.
On Windows GPU-Z is a lightweight utility designed to provide information about video cards and GPUs. It is more reliable then Windows Task Manager.