Theres no HA, so storage on one node is not avilable while that node is being rebooted. If you really wanted to you could evacuate a node with vol moves, move epsilon away from it, then reboot it to do the upgrade, then repeat for the rest of the nodes in your sim cluster. Aside from being an interesting academic excercise its probably not worth the effort. It would take a long time, require larger than defaul root volumes, and It wouldn't match the upgrade procedure on hardware anyway.
If this post resolved your issue, help others by selecting ACCEPT AS SOLUTION or adding a KUDO.