They're performing the resets when the cluster load goes down and they have sufficient capacity to handle the reset.
Capacity is one thing but inference still costs money.
alt Hacker News
Capacity is one thing but inference still costs money.