lecture 14: congestion control - computer science · 2013-02-26 · congestion control...

CSE 222A: Computer Communication Networks Alex C. Snoeren

Lecture 14:Congestion Control"

Thanks: Amin Vahdat, Dina Katabi

Lecture 14 Overview"  TCP congestion control review

  XCP Overview

2 CSE 222A – Lecture 14: Congestion Control"

Congestion Control Overview"  Challenge: how do we efficiently share network

resources among billions of hosts? ◆  Today: TCP

Hosts adjust rate based on packet losses ◆  Alternative solutions

Fair queuing, RED (router support) Vegas, packet pair (add functionality to TCP) Rate control, credits

3 CSE 222A – Lecture 14: Congestion Control"

Queuing Disciplines"  How to distribute buffers among users/flows

◆  When buffer overflows, which packet to drop?

  Simple solution: FIFO ◆  First in, first out ◆  If packet comes along with no available buffer space, drop it

4 CSE 222A – Lecture 14: Congestion Control"

Fair Queuing"  Goals:

◆  Allocate resources equally among all users/flows ◆  Low delay for interactive users ◆  Protection against misbehaving users

  Approach: simulate general processor sharing (from OS world) ◆  Bitwise round robin ◆  Need to compute number of competing flows at each instant

5 CSE 222A – Lecture 14: Congestion Control"

TCP Congestion Problems"  Original TCP sent full window of data

  When links become loaded, queues fill up, leading to: ◆  Congestion collapse: when round-trip time exceeds retransmit

interval this can create a stable condition in which every packet is being retransmitted many times

◆  Synchronized behavior: network oscillates between loaded and unloaded

»  Feedback loop

6 CSE 222A – Lecture 14: Congestion Control"

Jacobson’s Solution"  Transport protocols should obey conservation of

packets ◆  Use ACKs to clock injection of new packets

  Modify retransmission timer to adapt to variations in delay

  Infer network bandwidth from packet loss ◆  Drops è congestion è reduce rate ◆  No drops è no congestion è increase rate

  Limit send rate based on minimum of congestion window and advertised window

7 CSE 222A – Lecture 14: Congestion Control"

Tracking Bottleneck Bandwidth"  Throughput = window size/RTT

  Multiplicative decrease ◆  Timeout è dropped packet è cut window size in half

  Additive increase ◆  ACK arrives è no drop è increase window size by one

packet/window

8 CSE 222A – Lecture 14: Congestion Control"

TCP “Sawtooth”"  Oscillates around bottleneck bandwidth

◆  Adjusts to changes in competing traffic

0!

2!

4!

6!

8!

10!

12!

14!

16!

18!

window !(in segs)!

round-trip times!

Additive Increase/Multiplicative Decrease!

9 CSE 222A – Lecture 14: Congestion Control"

Slow Start"  How do we find bottleneck bandwidth?   Cannot use ACKs to clock without reaching equilibrium

◆  Start by sending a single packet Start slow to avoid overwhelming network

◆  Multiplicative increase until get packet loss Quickly find bottleneck Cut rate by half

◆  Shift into linear increase/multiplicative decrease

10 CSE 222A – Lecture 14: Congestion Control"

Slow Start"  Quickly find the bottleneck bandwidth

0!

50!

100!

150!

200!

250!

300!

0! 1! 2! 3! 4! 5! 6! 7! 8!

window !(in segs)!

round-trip times!

Slow Start!

11 CSE 222A – Lecture 14: Congestion Control"

Slow Start Problems"  Slow start usually overshoots bottleneck

◆  Leading to many lost packets in window ◆  Can lose up to half of window size

  Bursty traffic source ◆  Will cause bursty losses for other flows

  Short flows ◆  Can spend entire time in slow start

Especially for large bottleneck bandwidth

  Consider repeated connections to the same server ◆  E.g., for web connections

12 CSE 222A – Lecture 14: Congestion Control"

ACK Pacing in TCP"

  ACKs open up slots in the congestion/advertised window ◆  Bottleneck link determines rate to send ◆  ACK indicates one packet has left the network

13 CSE 222A – Lecture 14: Congestion Control"

Problems with ACK Pacing"  ACK compression

◆  Variations in queuing delays on return path changes spacing between ACKs

◆  Example: ACK waits for single long packet ◆  Worse with bursty cross-traffic

  What happens after a timeout? ◆  Potentially, no ACKs to time packet transmissions

  Congestion avoidance ◆  Slow start back to last successful rate ◆  Back to linear increase/multiplicative increase at this point

14 CSE 222A – Lecture 14: Congestion Control"

Two TCP Connections"  Reach equilibrium independent of initial bandwidth

(assuming equal RTTs)

0

2

4

6

8

10

12

14

16

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

window (in segs)

round-trip times

15 CSE 222A – Lecture 14: Congestion Control"

TCP “Friendliness”"  Problem: many different TCP implementations   If cut back more slowly after drops è grab bigger

share   If add more quickly after ACKs è grab bigger share   Incentive to cause congestion collapse

◆  Many TCP “accelerators” ◆  Easy to improve perf at expense of network

  Solutions? ◆  Per-flow fair queuing at router

16 CSE 222A – Lecture 14: Congestion Control"

Example: In TCP, Additive-Increase Multiplicative-Decrease (AIMD) controls both

Coupled because a single mechanism controls both

XCP argues decoupling solves the problem:

1.  To control congestion: use MIMD which shows fast response

2.  To control fairness: use AIMD which converges to fairness

Fairness vs. Efficiency"

17 CSE 222A – Lecture 14: Congestion Control"

XCP Advantages"  Tighter Congestion Control

◆  Small queues ◆  Almost no drops

  Scalable (no per-flow state)

  Flexible fairness definitions ◆  Max/min ◆  Proportional ◆  Differential bandwidth ◆  Etc.

18 CSE 222A – Lecture 14: Congestion Control"

1.  Congestion Controller 2.  Fairness Controller

XCP Overview"

19 CSE 222A – Lecture 14: Congestion Control"

Feedback

Round Trip Time

Congestion Window

Congestion Header

Feedback

Round Trip Time

Congestion Window

Feedback = + 0.1 packet

XCP Example"

20 CSE 222A – Lecture 14: Congestion Control"

Feedback = + 0.1 packet

Round Trip Time

Congestion Window

Feedback = - 0.3 packet

XCP Example"

21 CSE 222A – Lecture 14: Congestion Control"

Congestion Window = Congestion Window + Feedback

Routers compute feedback without any per-flow state

XCP Example"

22 CSE 222A – Lecture 14: Congestion Control"

Congestion Controller Fairness Controller Goal: Divides Δ between flows to converge to fairness Looks at a flow’s state in Congestion Header

Algorithm:

If Δ > 0 ⇒ Divide Δ equally between flows If Δ < 0 ⇒ Divide Δ between flows proportionally to their current rates

MIMD AIMD

Goal: Matches input traffic to link capacity & drains the queue Looks at aggregate traffic & queue Algorithm: Aggregate traffic changes by Δ Δ ~ Spare Bandwidth Δ  ~ - Queue Size So, Δ = α davg Spare - β Queue

Feedback Computation"

23 CSE 222A – Lecture 14: Congestion Control"

Bottleneck S1 S2

R1, R2, …, Rn

Sn

Dumbbell Topology"

24 CSE 222A – Lecture 14: Congestion Control"

Bottleneck Bandwidth (Mb/s)

Avg.

Util

izat

ion

Round Trip Delay (sec)

Avg.

Util

izat

ion

Efficiency ""

25 CSE 222A – Lecture 14: Congestion Control"

Start

40 Flows Stop the 40 Flows

Response to Dynamics"

26 CSE 222A – Lecture 14: Congestion Control"

Average Utilization

Average Queue

Drops

Short Flows"

27 CSE 222A – Lecture 14: Congestion Control"

Different RTT Same RTT

Avg.

Thr

ough

put

Avg.

Thr

ough

put

Fairness"

28 CSE 222A – Lecture 14: Congestion Control"

For Next Class…"   Read and review DC-TCP paper

  We’ll be in touch regarding Checkpoint meetings ◆  Email 1-2 page summary to Siva if you haven’t already

29 29 CSE 222A – Lecture 14: Congestion Control"