Hello Aaron
I’ll try to address your concerns so that you can have an idea of how to proceed from here.
This is how policy maps are set up. You cannot configure both priority and bandwidth, but either one. However, I believe what you are looking for is to configure a maximum allowed bandwidth for the priority queue. That can indeed be configured simultaneously and it is achieved using the policing feature.
Consider looking at the 4-class QoS strategy, as that provides a more suitable scenario for your situation, since you don’t need so many QoS classes.
OK there are a couple of issues here. If your voice is marked with EF, and your interactive video is also sometimes marked with EF, then you must find a way to differentiate them. This can be done either by matching an access list for particular ports (rather than QoS markings), or by changing the QoS markings on those particular packets. More on the latter later.
You can also use what is known as multi-level priority queues, which are further described in the following Cisco documentation. It may however not be supported on your platform:
Now does that mean that only 23% will be used? There is what is known as an implicit policer implemented in these cases, so yes. Take a look at this post which may give you some more information about this:
If you create a policy map that identifies the Zoom traffic (using IP address or port numbers for example) and place it incoming on the appropriate interface, you can use that policy map to change the QoS marking to whatever you like. Then they will reach the EF class map with the appropriate marking and you can treat them differently from your other traffic since you can now differentiate between them.
For most of the other questions, take a look at the following lesson:
I hope this has been helpful!
Laz