I’m just picturing Lakitu as a gaming systems operator picking the sever up out of the lake and putting it back on the track facing the right direction.
Lots of other folks covered a lot of the details, but I just want to point out that the answer really depends on the architecture of the system that the game runs inside of. Cloud native architecture has component failure as an expected characteristic and it is planned for, which means that architecture avoids unplanned outages better than legacy architectures. Even then, sometimes you end up with a component where you have a choice of investing thousands of engineering hours to avoid downtime during an upgrade, or you just tolerate a few minutes of downtime and spend those thousands of engineering hours on something with a better ROI.
I’m just picturing Lakitu as a gaming systems operator picking the sever up out of the lake and putting it back on the track facing the right direction.
Lots of other folks covered a lot of the details, but I just want to point out that the answer really depends on the architecture of the system that the game runs inside of. Cloud native architecture has component failure as an expected characteristic and it is planned for, which means that architecture avoids unplanned outages better than legacy architectures. Even then, sometimes you end up with a component where you have a choice of investing thousands of engineering hours to avoid downtime during an upgrade, or you just tolerate a few minutes of downtime and spend those thousands of engineering hours on something with a better ROI.