CSIT560 Internet Infrastructure: Switches and Routers

1
CSIT560Internet InfrastructureSwitches and
Routers

• Active Queue Management

Presented By Gary Po, Henry Hui and Kenny Chong
2
Agenda

• Why AQM?
• In Considerations
• AQM Algorithms
• Commercial Effort
• Conclusions

3
Why AQM?

• Two Classes of Router Algorithms for Congestion
  Control
• What is Congestion?
• Congestion avoidance in TCP, is it good enough?
• Our Choice
• Active Queue Management

4
What do we consider when implementing AQM?

• QoS
• Keep Average queue size small
• Bounded Delay
• Link Utilization
• Avoid Global Synchronization
• Absorbs bursts without dropping packets
• Fairness
• Punishes misbehaving flows
• Prevent bias against bursty connections
• Implementation
• Ease of Configurations
• Buffer Size Requirement (Large or Small)
• Per-flow State Information
• Computational Overhead

5
AQM Algorithms
6
FIFO Drop Tail

• Problems
• No isolation
• No policing
• Large queues for high utilizations
• Synchronization problem
• Lock-out problem

7
RED (Random Early Detection)

• FIFO scheduling

Define Two Threshold Values
Make Use of Average Queue Length
Case 1 Average Queue Length lt Min. Thresh Value
Admit the New Packet
8
RED (Contd)
Min thresh
Max thresh
Average queue length
Case 2 Average Queue Length between Min. and
Max. Threshold Value
Admit the New Packet With Probability p
Or Drop the New Packet With Probability 1-p
9
RED (Contd)
Min thresh
Max thresh
Case 3 Avg. Queue Length gt Max. Threshold Value
New Packet will be dropped
As no new packets can be admitted, the average
queue length decreases.
Until the average queue length drops below the
max threshold value
New packet could be admitted with a probability
p
or being dropped with a probability 1-p
10
RED Flow Diagram
11
RED (Contd)

• Queue Size versus Time

RED Queue Size
Global Synchronization solved
12
Unfairness of RED
An unresponsive flow occupies over 95 of
bandwidth
Unresponsive Flow (such as UDP)
32 TCP Flows 1 UDP Flow
13
CHOKe(CHOose and Keep)

• Based on RED
• Simple
• Designed for fairness
• Penalize the unresponsive flow

14
CHOKe (Contd)

• Mechanism

15
CHOKe (Contd)
Case 1 Average Queue Length lt Min. Thresh Value
Admit the New Packet
16
CHOKe (Contd)
Min thresh
Max thresh
Case 2 Avg. Queue Length is between Min. and
Max. Threshold Values
A packet is randomly chosen from the queue to
compare with the new arrival packet
If they are from different flows, the same logic
in RED applies
If they are from the same flow, both packets
will be dropped
17
CHOKe (Contd)
Min thresh
Max thresh
Case 3 Avg. Queue Length gt Max. Threshold Value
A random packet will be chosen for comparison
If they are from different flows, the new packet
will be dropped
If they are from the same flow, both packets
will be dropped
18
Evaluate CHOKes performance using NS-2
19
Simulation Scenario
source
destination
10Mbps
10Mbps
UDP
UDP
1Mbps
router
router
TCP
TCP

• Topology Dumb-bell
• Metrics throughput and queue size

20
Performance of CHOKe
Unresponsive Flow (UDP)
Fair Share Level
Bandwidth is evenly shared
32 TCP Flows 1 UDP Flow
21
Parameters

• Number of responsive/unresponsive flows
• Transfer rate of different flows
• Number of random candidates chosen for comparison

22
CHOKe Simulation

• Different Parameters, different performance

CHOKe-1 32 TCPs 1 UDP
CHOKe-2 32 TCPs, 3 UDPs of different rate
CHOKe-2 32 TCPs, 3 UDPs of same rate
CHOKe-2 32 TCPs, 1 UDP of high rate
23
Evolutions of AQM Algorithms

• RED
• Merits
• Early congestion detection
• No bias against bursty traffic
• No global synchronization
• Drawbacks
• Difficulty in parameter setting
• Insensitivity to traffic load and drain rates

• SRED
• Merits
• Stabilized queue occupancy
• Protection from misbehaving flows
• Drawbacks
• Some per-flow state (zombie list)
• RED disadvantages

• FRED
• Merits
• Good protection from misbehaving flows
• Drawbacks
• Per-flow state
• RED disadvantages

• BLUE
• Merits
• Simplicity
• High throughput
• Drawbacks
• No early congestion detection (Pdrop updated only
  on queue overflow or link idle events)
• Slow response and dependence on history

• REM
• Merits
• Low delay and small queues
• Independence of the number of users
• Drawbacks
• Some complexity due to parameters
• Low throughput for Web traffic
• Inconsistency with TCP sender mechanism works
  best with ECN

• LDC
• Merits
• Sensitivity to traffic load and drain rate
• Low delay
• Target delay achieved
• Intuitive parameters, meaningful to users (target
  delay)
• Drawbacks
• Some complexity due to parameters
• Low throughput in some cases

REM, AVQ, PI Controller
FIFO DropTail
BLUE
SFB
CHOKe
SAC
RED
SRED
FRED
24
Commercial Efforts Conclusion
25
Commercial Efforts Conclusion (Contd)

• Applying AQM over 3G wireless network a paper
  supported by Motorola Canada Ltd. (Mar. 2003)
• 3G network, real-time applications have hard time
  deadlines for packet delivery at the receiver.
• Use AQM to avoid long queuing delay and prevent
  expiring packets.

26
Commercial Efforts Conclusion (Contd)

• AQM improves overall system performance by
  increasing throughput and reducing end-to-end
  delay.

27
Commercial Efforts Conclusion (Contd)

• Effect of AQM on Web Performance a paper
  supported by Cisco Systems and IBM. (Aug. 2003)
• Proportional Integrator (PI) controller
• Random Exponential Marking (REM) controller
• Adaptive Random Early Detection (ARED).
• IETF proposed standard
• Explicit Congestion Notification (ECN)

28
Commercial Efforts Conclusion (Contd)

• ECN has significant impact with AQM scheme in web
  performance.
• Many researches and efforts are going on in the
  field of AQM.
• Simple and Easy to implement