Comment 3 for bug 1912460

Packet per second is a very general network performance metric. Like bandwidth, it is usually used to evaluate the packet forwarding performance of a device. But for a cloud, this will become a attack point. Your ovs, ovn or any other L2 drivers may get extremly high usage of CPU when user send small packet (topically 64B small) from the VM to others even the device is in lower bandwidth QoS limitation. Then your host services, other users' VMs will be under a higher failuer point. This is the main reason. For a networking developer, IMO, these background should be clear to you.

And this can be used for network quality assurance. The resource consumption of the system is determined according to the user's VM specifications. With the PPS limitation, the VM will not consume more CPU with smaller bandwidth.

@Rodolfo
For the question about how to implement this in agent side, IMO, the meter [1] can be used to achive the goal for OVN or OVS. Iptables can be used only for ml2/ovs.

[1] https://docs.pica8.com/pages/viewpage.action?pageId=5112993

And finally, for this RFE here, my original idea is to add the new rule type at first. For the real limitation drivers, IMO, we can start a new RFE for it.