Improving Adaptability and Fairness in Internet Congestion Control

1
Improving Adaptability and Fairness in Internet
Congestion Control

• May 30, 2001
• Seungwan Ryu
• PhD Student of IE Department
• University at Buffalo

2
I. Internet Congestion Control

• Internet Congestion Control
• Mathematical Modeling and Analysis
• Adaptive AQM and User Response
• Future Study Plan

3
I. Internet Congestion Control

• What is Congestion ?
• Congestion Control and Avoidance
• Implicit vs. Explicit feedback
• TCP Congestion Control
• Active Queue management (AQM)
• Explicit Congestion Notification (ECN)

4
What is congestion ?

• What is congestion ?
• The aggregate demand for bandwidth exceeds the
  available capacity of a link.
• What will be occur ?
• Performance Degradation
• Multiple packet loss
• Low link utilization (low Throughput)
• High queueing delay
• Congestion collapse

5
Congestion Control and Avoidance

• Two approaches for handling Congestion
• Congestion Control (Reactive)
• Play after the network is overloaded
• Congestion Avoidance (Proactive)
• Play before the network becomes overloaded

6
Implicit vs. Explicit feedback

• Implicit feedback Congestion Control
• Network drops packets when congestion occur
• 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)

7
Implicit vs. Explicit feedback - 2

• 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 complicate to implement
• Need to change both source and network algorithm
• Need cooperation between sources and network
  component

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
• cwnd min (pipe size, rwnd)
• self-clocking
• slow-start and congestion avoidance
• Examples
• TCP Tahoe, TCP Reno, TCP Vegas, etc.

9
TCP Congestion Control - 2

• Slow-start and Congestion Avoidance

cwnd
Congestion Avoidance
Slow Start
W1
W
4
2
1
RTT
RTT
Time
10
Active Queue Management (AQM) - 1

• Performance Degradation in current TCP Congestion
  Control
• Multiple packet loss
• Low link utilization
• Congestion collapse
• The role of the router (i.e., network)
• Control congestion effectively with a network
• Allocate bandwidth fairly

11
AQM - 2

• Problems with current router algorithm
• Use FIFO based tail-drop (TD) queue management
• Two drawbacks with TD lock-out, full-queue
• Possible solution AQM
• Drop packets before buffer becomes full
• Examples RED, BLUE, ARED, SRED, FRED,.
• Use (exponentially weighted) average queue length
  as an congestion indicator

12
AQM - 3

• Random Early Detection (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

13
RED
P
1
maxp
minth
maxth
K
14
Active Queue Management (AQM) - 4

• Problems with existing AQM Proposals
• Mismatch between macroscopic and microscopic
  behavior of queue length
• Insensitivity to the change of input traffic load
• Configuration (parameter setting) problem
• Reasons
• Queue length averaging
• use inappropriate congestion indicator
• Use inappropriate control function

15
Explicit Congestion Notification (ECN)

• Current congestion indication
• Use packet drop to indicate congestion
• source infer congestion implicitly
• ECN
• to give less packet drop and better performance
• use packet marking rather than drop
• need cooperation between sources and network
• need two bits in IP header ECT-bit, CE-bit

16
ECN - 2
ECT
CE
ECT
CE
1 0
1 1
IP Header
1
2
TCP Header
0
0
CWR
CWR
ACK TCP Header
1
ECN-Echo
3
TCP Header
1
4
CWR
Source
Router
Destination
17
Contents

• Internet Congestion Control
• Mathematical Modeling and Analysis
• Adaptive AQM and User Response
• Future Study Plan

18
II. Mathematical Modeling and Analysis

• An Overview
• Mathematical Modeling of AQM
• Window based packet switching and the Internet
• Mathematical modeling and analysis of AQM
• Problems with existing AQMs
• Problems with existing AQMs
• Adaptive congestion indicator and control function

19
Overview

• Goal of mathematical modeling
• see system dynamics (in steady state)
• capture main factors influence to performance
• provide design and/or operational recommendations
• Two approaches
• Modeling steady state TCP behaviors
• the square root law, PFTK
• assume TD queue management at the router
• Mathematical modeling and analysis of AQM (RED)

20
Overview - 2

• AQM modeling and analysis
• Analytic modeling and analysis
• Control Theoretic Analysis
• Window based modeling and Analysis
• Assumptions
• Poisson assumption for input traffic
• Fixed number of persistent TCP traffics
• Steady state window size saturation

21
Mathematical Modeling of AQM

• Window based packet switching Model (Yang 99)
• If link j is not congested
• If link j is congested

22
Mathematical Modeling of AQM - 2

• Window size of an individual connection
• Since
• Limitation of this model
• Assume infinite buffer size
• No buffer overflow
• No packet drop
• No queue management algorithm at routers

23
Mathematical Modeling of AQM - 3
24
Mathematical Modeling of AQM - 4

• Extend Yangs Model to AQM model
• Finite buffer capacity K
• The router use AQM to control congestion
• When congested
• Our Model
• Yangs Model

25
Mathematical Modeling of AQM - 5

• Case 1 Tail drop
• We obtain two relationship
• Finally, packet drop probability Pd

26
Mathematical Modeling of AQM - 6

• Case 2 AQM
• Let
• Then
• Packet drop prob. Pd

27
Mathematical Modeling of AQM - 7

• Congestion Indicator
• Input traffic load should be the congestion
  Indicator
• Current AQMs
• Use queue length Q as an alternative
• Assume that the input traffic load is fixed
  in equilibrium
• Reason
• can not measure(or estimate) exactly for on
  line implementation of packet drop function

28
Mathematical Modeling of AQM - 8

• Packet drop function
• Reason
• The traffic load fluctuate, NOT stay in
  equilibrium
• queue length is a function of input traffic
• Alternatively

29
Problems with existing AQMs

• Mismatch between macroscopic and microscopic
  behavior of queue length
• Insensitivity to the input traffic load variation
• parameter configuration problem

30
Problems with existing AQMs - 2

• Mismatch problem

31
Problems with existing AQMs - 3

• Mismatch between macroscopic and microscopic
  behavior of queue length

32
Problems with existing AQMs - 4

• Insensitivity to the input traffic load variation
• With light traffic (i.e., )

33
Problems with existing AQMs - 5

• Insensitivity to the input traffic load variation
• With medium traffic (i.e.,
  )

34
Problems with existing AQMs - 6

• Insensitivity to the input traffic load variation
• With heavy traffic (i.e.,
  )

35
Problems with existing AQMs - 7

• Parameter configuration problem
• Has been a main design issue since 1993
• many modified AQMs has been proposed
• Verified with simple simulation or simple
  experiment
• good for particular traffic conditions
• Real traffic is totally different.
• Need adaptive congestion indicator and control
  function
• Adaptive to input traffic load variation
• Avoid congestion NOT based on current state
  (i,e,. Q)

36
Contents

• Internet Congestion Control
• Mathematical Modeling and Analysis
• Adaptive AQM and User Response
• Future Study Plan

37
III. Adaptive AQM and User Response

• Input traffic load Prediction
• Adaptive AQM algorithms
• Adaptive parameter configuration
• Adaptive User response algorithm

38
Input traffic load Prediction

• Consider time-slotted model
• Time is divided into unit time slots, ?t, t0,1,
• calculate parameters at the end of each slot
• estimate Qt1 to detect congestion proactively
• Predict from measured input traffic ?t-1,
  ?t of past two time slots
• Then, predict of next time slot ?t

39
Adaptive AQM algorithms

• Algorithm I E-RED and E-GRED
• Enhanced-RED
• E-GRED similar to E-RED

40
Adaptive AQM algorithms - 2

• Algorithm II
• Use both predicted traffic intensity and
  current buffer utilization ?tQt/K
• Possible algorithms
• Example
• If ?t is low and is high more penalty to
  incoming packets
• If ?t is high and is low more penalty on
  existing packets
• Only High penalty for both packets when ?t and
  are high

41
Adaptive AQM algorithms - 3

• Algorithm III E-BLUE
• BLUE Algorithm
• uses packet drops and link idle for adjusting
  packet drop probability
• Can not avoid some degree of performance
  degradation
• Enhancement
• Use Virtual lower/upper bound (VL, VU)
• Combine predicted queue length with BLUE
• Impose penalty according to the traffic situation
  ( , )

42
Adaptive AQM algorithms - 4

• E-BLUE
• If , then pd pd- ?
• Else if VL lt ltVU,
• Else ( gtVU)
• pdpd?

43
Adaptive parameter configuration

• Adaptive queue length sampling interval ?t
• Previous recommendations
• In 22, minimum RTT was recommended
• In 65, static and link speed independent value
  was recommended
• However, models of 22, 65 were assumed to have
  persistent fixed N TCP traffics
• Our recommendation
• The amount of incoming traffic fluctuate with
  time
• Adjust ?t according to the varying traffic
  situation
• (i.e., adjust ?t according to the amount of
  input traffic)

44
Adaptive parameter configuration - 2
Q
?(i-1)
Time
?(i2)
?(i1)
?i
45
Adaptive parameter configuration - 3

• Adaptive filtering weight wq
• In RED, wq was recommended with 0.02 for
  long-term (macroscopic) performance goal
• Fixed small value of wq shows problems
• Parameter setting problem
• Insensitivity of control function to the change
  of traffic
• Fairness problem impose penalty to innocent
  packets
• Need to have adaptive wq to the change of traffic
  load
• One possible method
• Set wq as a function of current queue
  utilization,
• e.g., wq ? Qt/C , 0 lt ? lt 1

46
Adaptive User response algorithm

• AQM need work with intelligent source response
  for better performance
• Enhanced-ECN
• If receive ECN feedback in ?(t-1)
• If No ECN feedback in ?t
• If received ACK gt 0
• Else
• Else, Continue usual response to ECN feedback
• Else, Continue TCP Congestion Avoidance

47
Contents

• Internet Congestion Control
• Mathematical Modeling and Analysis
• Adaptive AQM and User Response
• Future Study Plan

48
IV. Future Study Plan

• Future Study plan a schedule
• Mathematical Modeling and Analysis
• Stability and Control Dynamics
• Alternative Modeling
• Control Theoretic Consideration
• Simulation plan
• Traffics
• Performance Metrics

49
Future Study plan a schedule

• Documentation
• Mathematical Modeling and Analysis
• Simulation plan
• Performance Metrics

50
Mathematical Modeling and Analysis

• Since pf(?,q) ,
• Then find equilibrium point (?,p)

p
?g(p)
Pf(?)
(?,p)
?
51
Mathematical Modeling and Analysis - 2

• Alternative Modeling
• State dependent service M/M/1 queueing model
• Lminth, KK-minth

52
Mathematical Modeling and Analysis - 3

• Service rates
• Steady state probabilities

53
Mathematical Modeling and Analysis - 3

• Control Theoretic Consideration

54
Simulation plan

• Goal of simulation study
• See dynamics and performance of our AQM
• Compare results with other AQM such as RED
• Use realistic traffic
• previous studies has been done with simple and
  unreal traffic (fixed number of persistent TCPs)
• Generate realistic Internet traffic
• Long-lived (FTP) and short-lived (web-like) TCP
  traffic
• UDP traffic CBR and/or ON/OFF

55
Performance Metrics

• TCP traffics
• Network-centric for aggregate traffic
• Throughput (or goodput)
• Packet dropping (marking) probability
• Link utilization (or queueing delay)
• User-centric for Individual traffic
• goodput (or throughput)
• mean response time (RTT)
• UDP traffic
• individual packet drop probability and its
  distribution