Does anybody know why our Cosmos DB is not scaling high enough even though Normalized RU is continuously at 100% for a couple of hours.

According to the documentation auto-scale should have increased provisioned throughput:

When you use autoscale, Azure Cosmos DB only scales the RU/s to the maximum throughput when the normalized RU consumption is 100% for a sustained, continuous period of time in a 5 second interval.

I’m guessing that the reason has to do with uneven partition RU, but I need a proper explanation instead of a guess:

Thanks