Congestion Control An Overview

1
Congestion Control An Overview

• -Jyothi Guntaka

2
Congestion

• What is congestion ?
• The aggregate demand for network resources
  exceeds the available capacity of a link.
• Effects of congestion
• Performance Degradation
• Multiple packet loss
• Low link utilization (low throughput)
• High queueing delay
• Congestion collapse

3
Approaches

• Congestion Avoidance (Proactive)
• Action before the network becomes overloaded.
• Congestion Control (Reactive)
• Action after the network is overloaded.

4
Implicit vs. Explicit feedback

• Implicit feedback Congestion Control
• Network drops packets when congestion occurs..
• Source infer congestion implicitly.
• time-out, duplicated ACKs, etc.
• Example end-to-end TCP congestion Control.
• Simple to implement but inaccurate.
• implemented only at Transport layer (e.g., TCP)

5
Implicit vs. Explicit feedback (contd)

• Explicit feedback Congestion Control
• Network component (e.g., router) provides
  congestion indication explicitly to sources.
• use packet marking, or RM cells (in ATM ABR
  control)
• Examples DECbit, ECN, ATM ABR CC, etc.
• Provide more accurate information to sources .
• But is more complicated to implement.
• Need to change both source and network algorithm
• Need cooperation between sources and network
  component

6
Importance of Congestion Control - Issues

• Fairness
• No universal (mathematical) definition for
  fairness.
• Depends on how many relevant dependencies are
  included in the model.
• All being equal concept (in best effort networks)
• Everybody gets equal service.
• All resources available to everybody.
• Each is expected to respect others and behave
  accordingly.
• When a new connection is added, everybody gets a
  little bit worse service.

7
Issues (contd)

• Economical aspect (in QoS enabled networks)
• You should get what you pay for.
• Old flows should not experience harm if a new
  flow is accepted.

8
TCP Congestion Control

• Use end-to-end congestion control
• use implicit feedback.
• e.g., time-out, triple duplicated ACKs, etc.
• use window based flow control.
• self-clocking
• slow-start and congestion avoidance
• Examples
• TCP Tahoe, TCP Reno, TCP Vegas, etc.

9
AIMD

• Additive Increase/Multiplicative Decrease (AIMD)
• Objective adjust to changes in the available
  capacity.
• New state variable per connection
  CongestionWindow.
• Limits how much data source has in transit.
• TCP source sending no faster than the slowest
  component. (network or destination host) can
  tolerate.
• Idea
• Increase CongestionWindow when congestion goes
  down.
• Decrease CongestionWindow when congestion goes up.

10
Random Early Detection (RED)

• RED
• Use network algorithm to detect incipient
  congestion.
• Design goals
• minimize packet loss and queueing delay
• avoid global synchronization
• maintain high link utilization
• removing bias against bursty source
• Achieve goals by
• randomized packet drop
• queue length averaging

11
Explicit Congestion Notification (ECN)

• Current congestion indication
• use packet drop to indicate congestion.
• source infers congestion implicitly.
• ECN
• to give less packet drop and better performance.
• use packet marking rather than drop.
• need cooperation between sources and network.

12
High Bandwidth-Delay Product Environments

• TCP congestion control performs poorly as
  bandwidth or delay increases.
• Proposed an eXplicit Control Protocol (XCP).
• Small queues
• Almost no drops
• Improved fairness
• Scalable (no per-flow state)

13
TCP vs. XCP

• TCP
• AIMD controls both congestion and fairness.
• XCP
• To control congestion MIMD is used which shows
  fast response.
• To control fairness AIMD is used which converges
  to fairness.
• XCP provides a joint design of end-systems and
  routers and it can co-exist with TCP.

14
TCP friendly congestion control

• TCP friendly a protocol that behaves like TCP
• Backs off if congestion and uses a fair share of
  resources.
• Protocol that obeys TCP long term throughput
  relation.
• Internet requirement new transport protocols
  must be TCP friendly
• Backs off if congestion and uses a fair share of
  resources.
• Applies also to application layer protocols
  transmitting over UDP, e.g., real time telephony
  or streaming applications.
• Rate control implemented on top of UDP as part of
  application.

15
TCP friendly congestion control (contd)

• Non-TCP friendly
• A protocol that takes more than its fair share of
  bandwidth (greedy).
• May cause fluctuations in network load and result
  in congestion collapse.
• How to protect your protocol against non-TCP
  friendly greedy protocols?
• RED is designed to solve this problem to some
  extent.

16
Research

• Develop a multipath protocol dealing with
  congestion control.
• Further study on measuring available bandwidth
  and other network path characteristics.
• Congestion control in wireless networks.
• Sensor networks.
• Completely wireless and hybrid wireless networks.

17
Acknowlegements

• Some of the slides from
• Improving Adaptability and Fairness in Internet
  Congestion Control.
• Congestion Control for High Bandwidth-Delay
  Product Environments.
• Lecture notes on Congestion Control
  (http//www.tct.hut.fi/opetus/s38188/2003/lect/188
  lect8net.pdf).