Congestion Control An Overview Jyothi Guntaka Congestion What

Congestion Control An Overview -Jyothi Guntaka

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

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

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

Implicit vs. Explicit feedback (…contd) Ø Explicit feedback Congestion Control u Network component (e. g. , router) provides congestion indication explicitly to sources. • use packet marking, or RM cells (in ATM ABR control) u u u 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

Importance of Congestion Control - Issues Ø Fairness u u Ø 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) u u 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.

Issues (…contd) Ø Economical aspect (in Qo. S enabled networks) u u You should get what you pay for. Old flows should not experience harm if a new flow is accepted.

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

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

Random Early Detection (RED) Ø RED u u Use network algorithm to detect incipient congestion. Design goals: • • u 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

Explicit Congestion Notification (ECN) Ø Current congestion indication u u Ø use packet drop to indicate congestion. source infers congestion implicitly. ECN u u u to give less packet drop and better performance. use packet marking rather than drop. need cooperation between sources and network.

High Bandwidth-Delay Product Environments Ø Ø TCP congestion control performs poorly as bandwidth or delay increases. Proposed an e. Xplicit Control Protocol (XCP). u u Small queues Almost no drops Improved fairness Scalable (no per-flow state)

TCP vs. XCP Ø TCP u Ø AIMD controls both congestion and fairness. XCP u u u 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.

TCP friendly congestion control Ø TCP friendly: a protocol that behaves like TCP u u Ø 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 u u u 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.

TCP friendly congestion control (…contd) Ø Non-TCP friendly: u u Ø 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? u RED is designed to solve this problem to some extent.

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

Acknowlegements Ø Some of the slides from u u u Improving Adaptability and Fairness in Internet Congestion Control for High Bandwidth-Delay Product Environments. Lecture notes on Congestion Control (http: //www. tct. hut. fi/opetus/s 38188/2003/le ct/188 lect 8 net. pdf).