When configuring WFQ, what is the default number of dynamic queues based on?

A.
hold-queue limit

B.
congestive discard threshold (CDT)

C.
interface bandwidth

D.
hash of the packet headers

E.
inter-packet arrival rate

F.
drop probability denominator

Explanation:

Weighted Fair Queuing (WFQ) classifies traffic entering the queue based on traffic flows. The actual classification can be based on source and destination addresses, the protocol and TCP port numbers. Each flow is given its own queue. In its simplest form WFQ services each of these queues on a round robin basis. This means that every flow of traffic has an equal share of the available bandwidth, if it is required. Hence the term "fair" queue. The benefit for low volume traffic is reduced and predictable latency. For many applications this default behavior of WFQ is sufficient, however, some applications need specific QoS guarantees that require more than simply "fair" access to the bandwidth. In this case, the "weight" needs to be modified so that WFQ does not share bandwidth on a round-robin basis, but is influenced by the class or priority of the traffic in the flow.
Weighted fair queuing is activated on an interface using the fair-queue command:

Router(config)# interface serial 0/0
Router(config-if)# fair-queue

Note that the queuing strategies described here apply to packets at Layer 3. Consequently, these techniques can only be used on router interfaces. Physical interfaces on the Catalyst 3550 and Catalyst 6500 switches are used as router interfaces after applying the command:

switch(config-if)# no switchport

These features are not available on regular Layer 2 switch ports and none of these features are available on Layer 2 switches such as the Catalyst 2950.