H3C Technologies H3C S7500E Series Switches - Manual (Page 71)

7-1
7
Congestion Avoidance
When configuring congestion avoidance, go to these sections for information you are interested in:
z Congestion Avoidance Overview
z Introduction to WRED Configuration
z Configuring WRED on an Interface
z Displaying and Maintaining WRED
Congestion Avoidance Overview
Avoiding congestion before it occurs to deteriorate network performance is a proactive approach to
improving network performance. As a flow control mechanism, congestion avoidance actively drops
packets when congestion is expected to occur or deteriorate by monitoring the utilization of network
resources (such as queues or memory buffers) to alleviate the load on the network.
Compared with end-to-end flow control, this flow control mechanism controls the load of more flows in
a device. When dropping packets from a source end, it cooperates with the flow control mechanism
(such as TCP flow control) at the source end to regulate the network traffic size. The combination of
the local packet drop policy and the source-end flow control mechanism helps maximize throughput
and network use efficiency and minimize packet loss and delay.
Traditional packet drop policy
Tail drop is the traditional approach to congestion avoidance. In this approach, when the size of a
queue reaches the maximum threshold, all the subsequent packets are dropped.
This results in global TCP synchronization. That is, if packets from multiple TCP connections are
dropped, these TCP connections go into the state of congestion avoidance and slow start to reduce
traffic, but traffic peak occurs later. Consequently, the network traffic jitters all the time.
RED and WRED
You can use random early detection (RED) or weighted random early detection (WRED) to avoid
global TCP synchronization.
Both RED and WRED avoid global TCP synchronization by randomly dropping packets. Thus, while
the sending rates of some TCP sessions slow down after their packets are dropped, other TCP
sessions remain at high sending rates. As there are always TCP sessions at high sending rates, link
bandwidth is efficiently utilized.
The RED or WRED algorithm sets an upper threshold and lower threshold for each queue, and
processes the packets in a queue as follows:
z When the queue size is shorter than the lower threshold, no packet is dropped;
z When the queue size reaches the upper threshold, all subsequent packets are dropped;
z When the queue size is between the lower threshold and the upper threshold, the received
packets are dropped at random. The longer a queue is, the higher the drop probability is. However,
a maximum drop probability exists.