1. @Yuan Xue (yuan.xue@vanderbilt.edu) A special acknowledge goes to J.F Kurose and K.W. Ross Some of the slides used in this lecture are adapted from their original slides that accompany the book “Computer Networking, A Top-Down Approach” All material copyright 1996-2009 J.F Kurose and K.W. Ross, All Rights Reserved CS 283Computer Networks Spring 2011 Instructor: Yuan Xue

2. @Yuan Xue (yuan.xue@vanderbilt.edu) Transport Layer Outline Overview 3.1 Transport-layer services 3.2 Multiplexing and demultiplexing 3.3 UDP: Connectionless transport 3.4 Principles of reliable data transfer 3.5 Connection-oriented transport: TCP Demultiplexing segment structure reliable data transfer flow control connection management Congestion control 3.6 Principles of congestion control 3.7 TCP congestion control

3. @Yuan Xue (yuan.xue@vanderbilt.edu) Principles of Congestion Control Congestion: informally: “too many sources sending too much data too fast for network to handle” different from flow control! manifestations: lost packets (buffer overflow at routers) long delays (queueing in router buffers) a top-10 problem!

4. @Yuan Xue (yuan.xue@vanderbilt.edu) Approaches towards congestion control end-end congestion control: no explicit feedback from network congestion inferred from end-system observed loss, 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 two broad approaches towards congestion control:

5. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP congestion control: r goal: TCP sender should transmit as fast as possible, but without congesting network m Q: how to find rate just below congestion level r decentralized: each TCP sender sets its own rate, based on implicit feedback: m ACK: segment received (a good thing!), network not congested, so increase sending rate m lost segment: assume loss due to congested network, so decrease sending rate

6. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP congestion control: bandwidth probing r “probing for bandwidth”: increase transmission rate on receipt of ACK, until eventually loss occurs, then decrease transmission rate m continue to increase on ACK, decrease on loss (since available bandwidth is changing, depending on other connections in network) ACKs being received, so increase rate X X X X X loss, so decrease rate sending rate time r Q: how fast to increase/decrease? m details to follow TCP’s “sawtooth” behavior

7. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP Congestion Control : Window-based sender limits rate by limiting number of unACKed bytes “in pipeline”: cwnd: differs from rwnd (how, why?) sender limited by min(cwnd,rwnd) roughly, cwnd is dynamic, function of perceived network congestion rate = cwnd RTT bytes/sec LastByteSent-LastByteAcked  cwnd cwnd bytes RTT ACK(s)

8. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP Congestion Control: Overview segment loss event: reducing cwnd timeout: no response from receiver cut cwnd to 1 3 duplicate ACKs: at least some segments getting through (recall fast retransmit) cut cwnd in half, less aggressively than on timeout ACK received: increase cwnd r slowstart phase: m increase exponentially fast (despite name) at connection start, or following timeout r congestion avoidance: m increase linearly

9. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP Slow Start when connection begins, cwnd = 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 increase rate exponentially until first loss event or when threshold reached double cwnd every RTT done by incrementing cwnd by 1 for every ACK received Host A one segment RTT Host B two segments four segments

10. @Yuan Xue (yuan.xue@vanderbilt.edu) Transitioning into/out of slowstart ssthresh: cwnd threshold maintained by TCP on loss event: set ssthresh to cwnd/2 remember (half of) TCP rate when congestion last occurred when cwnd >= ssthresh : transition from slowstart to congestion avoidance phase

11. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP: congestion avoidance when cwnd > ssthresh grow cwnd linearly increase cwnd by 1 MSS per RTT approach possible congestion slower than in slowstart implementation: cwnd = cwnd + MSS/cwnd for each ACK received r ACKs: increase cwnd by 1 MSS per RTT: additive increase r loss: cut cwnd in half (non-timeout-detected loss ): multiplicative decrease AIMD AIMD: Additive Increase Multiplicative Decrease

12. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP congestion control FSM: overview slow start congestion avoidance fast recovery cwnd > ssthresh loss: timeout loss: timeout new ACK loss: 3dupACK loss: 3dupACK loss: timeout

13. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP FSM: link congestion control with reliable delivery slow start congestion avoidance fast recovery timeout ssthresh = cwnd/2 cwnd = 1 MSS dupACKcount = 0 retransmit missing segment timeout ssthresh = cwnd/2 cwnd = 1 MSS dupACKcount = 0 retransmit missing segment  cwnd > ssthresh cwnd = cwnd+MSS dupACKcount = 0 transmit new segment(s),as allowed new ACK cwnd = cwnd + MSS (MSS/cwnd) dupACKcount = 0 transmit new segment(s),as allowed new ACK. dupACKcount++ duplicate ACK ssthresh= cwnd/2 cwnd = ssthresh + 3 retransmit missing segment dupACKcount == 3 dupACKcount++ duplicate ACK ssthresh= cwnd/2 cwnd = ssthresh + 3 retransmit missing segment dupACKcount == 3 timeout ssthresh = cwnd/2 cwnd = 1 dupACKcount = 0 retransmit missing segment cwnd = cwnd + MSS transmit new segment(s), as allowed duplicate ACK cwnd = ssthresh dupACKcount = 0 New ACK  cwnd = 1 MSS ssthresh = 64 KB dupACKcount = 0

14. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP congestion control FSM: overview slow start congestion avoidance fast recovery cwnd > ssthresh loss: timeout loss: timeout new ACK loss: 3dupACK loss: 3dupACK loss: timeout

15. @Yuan Xue (yuan.xue@vanderbilt.edu) Summary: TCP Congestion Control when cwnd < ssthresh, sender in slow-start phase, window grows exponentially. when cwnd >= ssthresh, sender is in congestion- avoidance phase, window grows linearly. when triple duplicate ACK occurs, ssthresh set to cwnd/2, cwnd set to ~ ssthresh when timeout occurs, ssthresh set to cwnd/2, cwnd set to 1 MSS.

16. @Yuan Xue (yuan.xue@vanderbilt.edu) More on TCP Congestion Control Q: Why AIMD? Can I come up with an alternative design for TCP? A: Yes. Actually TCP has different versions and different implementations. Reno, Tahoe, Vegas, etc

17. @Yuan Xue (yuan.xue@vanderbilt.edu) Popular “flavors” of TCP ssthresh TCP Tahoe TCP Reno Transmission round cwnd window size (in segments)

18. @Yuan Xue (yuan.xue@vanderbilt.edu) More on TCP Congestion Control Q: Is TCP good (enough? optimal?) A: Well… this is a long story… but the answer is no..

19. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP throughput Q: what’s average throughout of TCP as function of window size, RTT? ignoring slow start let W be window size when loss occurs. when window is W, throughput is W/RTT just after loss, window drops to W/2, throughput to W/2RTT. average throughout:.75 W/RTT

20. @Yuan Xue (yuan.xue@vanderbilt.edu) TCP Futures: TCP over “long, fat pipes” example: 1500 byte segments, 100ms RTT, want 10 Gbps throughput requires window size W = 83,333 in-flight segments throughput in terms of loss rate: new versions of TCP for high-speed

21. @Yuan Xue (yuan.xue@vanderbilt.edu) fairness goal: if K TCP sessions share same bottleneck link of bandwidth R, each should have average rate of R/K TCP connection 1 bottleneck router capacity R TCP connection 2 TCP Fairness

22. @Yuan Xue (yuan.xue@vanderbilt.edu) More on TCP Congestion Control Q: What aspect should I consider, if I am going to work on my “TCP” design? A: Throughput  Can it exploit the link capacity? Being aggressive may not bring additional benefit Fairness  Can it share the link capacity fairly among all flows? Friendliness  What if it runs with TCP? Experiment for testing your “TCP“ design: one flow; multiple flows; With background traffic (TCP, UDP)

23. @Yuan Xue (yuan.xue@vanderbilt.edu) More on TCP Congestion Control Q: Can I change the TCP implementation on my computer to speed up my Internet experience?! A: Sure. But the first thing to consider: Do you have the control of both sender and receiver? If you only implement the sender, what can you do? What if you only control the receiver?

24. @Yuan Xue (yuan.xue@vanderbilt.edu) Chapter 3: Summary principles behind transport layer services: multiplexing, demultiplexing reliable data transfer flow control congestion control instantiation and implementation in the Internet UDP TCP Next: leaving the network “edge” (application, transport layers) into the network “core”