Configuring Congestion Avoidance; Overview; Tail Drop; Red And Wred - HP MSR4080 Configuration Manual

Configuring congestion avoidance

Overview

Avoiding congestion before it occurs is a proactive approach to improving network perform
flow control mechanism, congestion avoidance:
• Actively monitors network resources (such as queues and memory buffers).
Drops packets when congestion is expected to occur or deteriorate.
•
When dropping packets from a source end, congestion avoidance cooperates wi
mechanism at the source end to regulate the network traffic size. The combination of
drop policy and the source-end flow control mechanism helps maximize throughp
efficiency and minimize packet loss and delay.

Tail drop

Congestion management techniques drop all packets that are arriving at a full queue
mechanism results in global TCP synchronization. If packets from multiple TCP connectio
these TCP connections go into the state of congestion avoidance and slow start to re
traffic peak occurs later. Consequently, the network traffic jitters all the time.

RED and WRED

Y ou 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. When the
sending rates of some TCP sessions slow dow
remain at high sending rates. Link bandwidth is e
rates always exist.
The RED or WRED algorithm sets an upper threshold
the packets in a queue as follows:
When the queue size is shorter than the lower threshold, no packet is dropped.
•
When the queue size reaches the upper threshold, all subsequent packets are dropped.
•
When the queue size i
•
are dropped at random. The drop probability in a queue inc
the maximum drop probability.
If the current queue size is
policy, burst traffic is not f
with the upper threshold and lower threshold to determine the drop probability.
Th e a verage queue size reflects the queue size change trend but is not sensitive to burst queue siz
changes, and burst traffic can be fairly treated.
n after their packets are dropped, other TCP sessions
s between the lower threshold and the upper threshold, the received packets
compared with the upper threshold and lower threshold to determine the drop
airly treated. To solve this problem, WRED compares the average queue size
62
fficiently used, because TCP sessions at high sending
and lower threshold for each queue, and processes
reases along with the queue size under
ance. As a
th the flow control
the local packet
ut and network use
. This tail drop
ns are dropped,
duce traffic, but
e