Congestion Management Configuration Approaches - HP 4800G Series Configuration Manual

Hide thumbs

Table of Contents

Short packets and long packets are fairly scheduled: if there are both long packets and short

packets in queues, statistically the short packets should be scheduled preferentially to reduce the

jitter between packets as a whole.

Compared with FQ, WFQ takes weights into account when determining the queue scheduling order.

Statistically, WFQ gives high priority traffic more scheduling opportunities than low priority traffic. WFQ

can automatically classify traffic according to the "session" information of traffic (protocol type, TCP or

UDP source/destination port numbers, source/destination IP addresses, IP precedence bits in the ToS

field, and so on), and try to provide as many queues as possible so that each traffic flow can be put into

these queues to balance the delay of every traffic flow as a whole. When dequeuing packets, WFQ

assigns the outgoing interface bandwidth to each traffic flow by precedence. The higher precedence

value a traffic flow has, the more bandwidth it gets.

Additionally, WFQ can work with the minimum guaranteed bandwidth mechanism. You can configure a

minimum guaranteed bandwidth for each WFQ queue to guarantee that each WFQ queue is

guaranteed of the bandwidth when congestion occurs. The assignable bandwidth (assignable

bandwidth = total bandwidth – the sum of the minimum guaranteed bandwidth for each queue) is

allocated to queues based on queue priority.

For example, assume that the total bandwidth of a port is 10 Mbps, and there are five flows on the port

currently, with the precedence being 0, 1, 2, 3, and 4 and the minimum guaranteed bandwidth being 128

kbps, 128 kbps, 128 kbps, 64 kbps, and 64 kbps respectively.

The assignable bandwidth = 10 Mbps – (128 kbps + 128 kbps + 128 kbps + 64 kbps + and 64 kbps)

The total assignable bandwidth quota is the sum of all the (precedence value + 1)s, that is, 1 + 2 +

3 + 4 + 5 = 15.

The bandwidth percentage assigned to each flow is (precedence value of the flow + 1)/total

assignable bandwidth quota. The bandwidth percentages for the flows are 1/15, 2/15, 3/15, 4/15,

and 5/15 respectively.

The bandwidth finally assigned to a queue = the minimum guaranteed bandwidth + the bandwidth

allocated to the queue from the assignable bandwidth

Because WFQ can balance delay and jitter among flows when congestion occurs, it is effectively

applied in some special occasions. For example, WFQ is used for the assured forwarding (AF) services

of the Resource Reservation Protocol (RSVP). In Generic Traffic Shaping (GTS), WFQ is used to

schedule buffered packets.

SP+WRR queuing

By assigning some queues on the port to the SP scheduling group and the others to the WRR

scheduling group (that is, group 1), you implement SP + WRR queue scheduling on the port. Packets in

the SP scheduling group are scheduled preferentially. When the SP scheduling group is empty, packets

in the WRR scheduling group are scheduled. Queues in the SP scheduling group are scheduled with

the SP queue scheduling algorithm. Queues in the WRR scheduling group are scheduled with WRR.