I have a Tesla V100 video card, tell me what means to monitor it, namely, to watch the utilization of graphics cores, GPU, the utilization of CUDA cores and it is desirable to log all this in Zabbix. I will be grateful for any information, thank you.
nvidia-smi is used to monitor gpus, zabbix plugin:
https://share.zabbix.com/cat-server-hardware/other/nvidia-smi-monitoring-for-multiple-gpus
Monitoring temperatures might be useful.
Thank you very much, this is what me need.