From my own perspective, the business case is reliability. We can take responsibility for the entire rack. No “actually that bug is in code from our vendor, we’

From my own perspective, the business case is reliability. We can take responsibility for the entire rack. No “actually that bug is in code from our vendor, we’ll file a ticket upstream” and then maybe someday it’ll get fixed and maybe someday eventually roll out. If (let’s be real, when: nothing is perfect…) problems in our system are found, we can fix them. There’s other reasons too. But imho that’s the big one.