As far as I know, there is a system of credits, which decides, how much cpu usage you can get. If you don't use a lot of you usage, then you can collect credits, until a max amount of credits, which you can use for more processor usage, if required. This depends on the size of you cloud container.
I hope someone of Mendix can verify my answer.
I'm not a 100% sure about this, since I don't know the internals of the Mendix Cloud. What we do know is that it is hosted on AWS. Searching for "CPU Credit Amazon" you will find the page CPU Credits and Baseline Performance for Burstable Performance Instances. I believe this page explains the system of CPU Credits well. I know that the Mendix Cloud allows your container to use some more CPU in short bursts, however I never linked it to the system described on the documentation page. Taking all into account is would say that Mendix is probably using the burstable performance instances on AWS. Below some relevant quotes:
Traditional Amazon EC2 instance types provide fixed performance, while burstable performance instances provide a baseline level of CPU performance with the ability to burst above that baseline level. The baseline performance and ability to burst are governed by CPU credits. A CPU credit provides the performance of a full CPU core for one minute.
I agree that this should be documented somewhere.
I just got an answer from Mendix.
This is the official aws documentation:
A Mendix M container DB is running on an RDS t2.small instance. Maybe that helps other people in the future.