Less Reboots - Most updates shouldn't need a full server reboot you should be able to restart just the service to apply an update
@Rich We have to take care about superseded patches, GDR, QFE patches and patching sequence. The way it is actually running is great and a global patch is really weird to implement because all OS installations have not the same roles and features installed.
Rich Siegel commented
thats one way. the other way is simplified update approach:
I mentioned this 1 on nano forum
use an active-passive root partition scheme to allow updating the entire OS as a single unit, instead of package by package. In this way the updates are quick, reliable and able to be easily rolled back. It is silly to distribute individual patches that take so long to install