1 / 21

CENG415 – Communication Networks

CENG415 – Communication Networks. Lectures 11 Principle of Congestion Control. Transport layer. We will cover Transport-layer services Multiplexing and demultiplexing Connectionless transport: UDP Principles of reliable data transfer Connection-oriented transport: TCP

moral
Download Presentation

CENG415 – Communication Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CENG415 – Communication Networks Lectures 11 Principle of Congestion Control

  2. Transport layer We will cover • Transport-layer services • Multiplexing and demultiplexing • Connectionless transport: UDP • Principles of reliable data transfer • Connection-oriented transport: TCP • segment structure • reliable data transfer • flow control • connection management • Principles of congestion control • TCP congestion control

  3. TCP overview Congestion: • Too many sources sending too much data too fast for network to handle • different from flow control (do not overload receiver) • Manifestations & Consequences: • lost packets (buffer overflow at routers) • long delays (queuing in router buffers)

  4. lout lin : original data unlimited shared output link buffers Host A Host B Causes / Costs of congestion: scenario 1 • Two senders, two receivers • one router, infinite buffers • no retransmission • λin: bytes/seconds average rate of arriving bytes from the application layer. • λout: bytes/seconds average rate of arriving bytes to the receiver transport layer.

  5. lout lin : original data unlimited shared output link buffers Host A Host B Causes / Costs of congestion: scenario 1 • λin These bytes arrive once to thetransport layer. If we have retransmission, the transport layer retransmit, the application layer has nothing to do with retransmission • λoutThese bytes include all retransmission bytes

  6. Causes / Costs of congestion: scenario 1 • C is the capacity of the link (number of bytes per second) • A is sending to be and B is sending to A • each is using C/2  Max value that can go through the link • In first graph, λoutcannot exceed C/2 • In second graph, when λinexceeds C/2, the buffering is unbounded and the waiting time is large

  7. Causes / Costs of congestion: scenario 2 • One router, finite buffer • Sender retransmission of lost packet • always:λin= λout (goodput) Host A lout lin : original data l'in: original data, plus retransmitted data Host B finite shared output link buffers

  8. Causes / Costs of congestion: scenario 2 Host A lout lin : original data • “perfect” retransmission only when loss: λ’in > λ out • retransmission of delayed (not lost) packet makes λ’in larger (than perfect case) for same λ out • “costs” of congestion: • more work (retransmissions) for given “goodput” • unneeded retransmissions: link carries multiple copies of pkt l'in: original data, plus retransmitted data Host B finite shared output link buffers

  9. Host A Host B Causes / Costs of congestion: scenario 3 • four senders • multihop (multi routers) paths • timeout/retransmit • Another “cost” of congestion: • when packet dropped, any “upstream transmission capacity used for that packet was wasted! lout lin : original data l'in: original data, plus retransmitted data finite shared output link buffers

  10. Approaches towards congestion control Two broad approaches towards congestion control: End-end congestion control: • no explicit feedback from network • congestion inferred from end-system observed loss and delay • approach taken by TCP Network-assisted congestion control: • routers provide feedback to end systems • single bit indicating congestion (SNA, DECbit, TCP/IP ECN, ATM) • explicit rate sender should send at

  11. Transport layer We will cover • Transport-layer services • Multiplexing and demultiplexing • Connectionless transport: UDP • Principles of reliable data transfer • Connection-oriented transport: TCP • segment structure • reliable data transfer • flow control • connection management • Principles of congestion control • TCP congestion control

  12. TCP congestion control Revision Windows: • Receiver Buffer: number of bytes that can be buffered • Receiver Window: number of bytes that can be received and buffered • Congestion window: defined by the sender according to the number of duplicate acknowledgements • What the sender can send is MIN(RcvBuffer, RcvWindow, CongWin)

  13. CongWin rate = Bytes/sec RTT TCP congestion control - Details • sender limits transmission according to Congestion: LastByteSent-LastByteAcked CongWin • CongWin is dynamic, function of perceived network congestion How does sender perceive congestion? • loss event = timeout or 3 duplicate acks • TCP sender reduces rate (CongWin) after loss event • two mechanisms: • AIMD (additive increase, multiplicative decrease) • slow start (after RTO)

  14. TCP congestion control - AIMD Additive increase / multiplicative decrease • Approach: increase transmission rate (window size), probing for usable bandwidth, until loss occurs • additive increase: increase CongWin by 1 MSS every RTT until loss detected • multiplicative decrease: cut CongWin in half after loss • Loss detected with 3 duplicate ACKs Lost packet (3 duplicate ACKs) congestion window size time

  15. TCP congestion control – Slow Start • When connection begins, CongWin = 1 MSS* • Example: MSS = 500 bytes & RTT = 200 msec • initial rate = 20 kbps • available bandwidth may be >> MSS/RTT • desirable to quickly ramp up to respectable rate  When connection begins, increase rate exponentially fast until first loss event (doubles every RTT) * MSS = maximum segment size (e.g.. , 1420 bytes). specified as a TCP option in SYN and SYN-ACK packets.

  16. time TCP congestion control – Slow Start Host B Host A • When connection begins, increase rate exponentially until first loss event: • double CongWin every RTT • done by incrementing CongWin for every ACK received • Summary: initial rate is slow but ramps up exponentially fast one segment RTT two segments four segments

  17. TCP congestion control – Slow Start • Implementation • Variable Threshold • At loss event, Threshold is set to 1/2 of CongWin just before loss event Refinement • When CongWin gets to 1/2 of its value before timeout, we switch to linear increase Congestion Window /mss “Slow Start” Transmission Round (time/rtt)

  18. Slow Start - Refinement After timeout event: • CongWin instead set to 1 MSS; • window then grows exponentially • to a threshold, then grows linearly After 3 dup ACKs: • CongWin is cut in half • window then grows linearly “Fast Recovery” • Philosophy: • 3 dup ACKs indicates network capable of delivering some segments • timeout indicates a “more alarming” congestion scenario Congestion Window /mss “Slow Start” Transmission Round (time/rtt)

  19. Scenario • CongWin <= Threshold: Doubles each RTT (add MSS for each MSS ACKed) • CongWin > Threshold: Adds MSS each RTT • Time Out: Threshold = 1/2 CongWin, CongWin = 1 (Slow-Start) • 3-Dup Ack: Threshold = 1/2 CongWin, CongWin = Threshold (Fast Recovery)

  20. TCP congestion control - Summary • CongWinis below Threshold, sender in slow-start phase,  window grows exponentially. • CongWinis above Threshold, sender is in congestion-avoidancephase Window grows linearly. • Three duplicate ACKoccurs Thresholdset to CongWin/2 and CongWin set to Threshold (no slow start, linear mode immediatly). • Timeout occurs Thresholdset to CongWin/2 and CongWin is set to 1 MSS.

  21. TCP sender congestion control

More Related