I have done setup on tesla v100 -16gb aws (ubantu 20.4) machine.
i have installed cuda 11.2 & cudnn . Everything is working fine. I was able to do training of my computer vision model
My server is on for 3 days .once the training is over i used linx command to shutdown server
sudo shutdown now
when i was checking same server next day, automatically cuda is corrupted.
May i know what is the reason for this?
Every time i shutdown machine . Next day again i need to do all setup.
i have attached screenshot. when i turn on my server its cuda corrupted.Thanks in advance, Im looking for solution to solve this problem.