I am trying to understand cisco QoS techniques. I came to know that queueing techniques is used(or comes alive) only when you experience conjestion on your network. Here I am a little confused with a few scenarios in front of me. I am trying to post them in the best way I can describe.
Suppose you have a 30 Mbps link. I have defined 3 classes of traffic
a. Voice - 50 % Bandwidth - Priority
b. Applications : 25% Bandwidth
c. Best Effort : remaining with WFQ
Case 1: Is queing techniques applies to the buffer of my interface here ? If so what is the size of the buffer - 30 Mbps ?
Case 2 : You are not expereincing conjestion at all. eg :- We are using only 15 Mbps of the link.
Do we have any sort queuing happening at this moment ?
Case 3 : You are expereincing congestion. Cause of the congestion is application traffic. As soon as the buffers start filling up, do queing limit my applications traffic to just 20 % of my link though there is no other traffic is alive at this moment ?
Please answer. If some one can share a good url to understand these as well is very much appreciated.