TCP Congestion Control:

Algorithms and Analysis

Simon S. Lam
Department of Computer Sciences
The University of Texas at Austin

Little’s Law

Average
g population
p p
= (average delay) x N
(throughput)
average delay  1  delayi
N i1
where N is number of departures
throughput  N/T
where T is duration of observation
average population (to be defined)

Try homework problem at

http://www.cs.utexas.edu/users/lam/cs356/homework/hw2.html
TCP Congestion Control (Simon Lam) 2

1
 Number in system n(t)

Time t

average population  1 T0 n(t)dt

T

TCP Congestion Control (Simon Lam) 3

Sliding Window Protocol

 Consider an infinite array, Source, at the
sender, and an infinite array, Sink, at the
receiver.
Source: send window

P1
0 1 2 a–1 a s–1 s
Sender
acknowledged unacknowledged

next expected r + RW – 1
Sink : received
P2
0 1 2 r
Receiver
delivered receive window

RW receive window size

SW send window size (s - a  SW) TCP Congestion Control (Simon Lam) 4

2
 Sliding Windows in Action
 Data unit r has just been received by P2
 Receive window slides forward

 P2
P sends
d cumulative
l ack
k with
h sequence
number it expects to receive next (r+3)
Source: send window

P1
0 1 2 a–1 a s–1 s
Sender
acknowledged unacknowledged

r+3

next expected r + RW – 1
Sink:
P2
0 1 2 r
Receiver
delivered receive window

TCP Congestion Control (Simon Lam) 5

Sliding Windows in Action

 P1 has just received cumulative ack with
r+3 as next expected sequence number
 Send window slides forward

Source: send window

P1
0 1 2 a–1 a s–1 s
Sender
acknowledged

next expected r + RW – 1
Sink:
P2
0 1 2 r
Receiver
delivered receive window

TCP Congestion Control (Simon Lam) 6

3
 Window Flow Control
RTT

Source 1 2 W 1 2 W
time

data ACKs

Destination 1 2 W 1 2 W
time

 ~ W packets per RTT when no loss

 Lost packet detected by missing ACK
(note: timeout value TO > RTT)

TCP Congestion Control (Simon Lam) 7

Throughput (send rate)

 Limit the number of unacked transmitted
packets in the network to window size W

W
 Throughput  packets/sec
RTT
W  MSS
= bytes/sec
RTT

 Where did we apply Little’s Law?

Answer: Consider send buffer

TCP Congestion Control (Simon Lam) 8

4
Clarifications
 Average number in the send buffer is typically
less than W unless packet arrival rate to send
buffer is infinite -> previous formula provides a
throughput upper bound

 If each packet may be lost with rate p, then the

average delay is
(1  p)  RTT  p  TO
Since TO > RTT, actual throughput is smaller.
 With loss, g
goodput
p is
(1  p )  throughput
th h t

 Note: in some papers and other context (e.g., random

access protocols), goodput is called throughput. To avoid
confusion, throughput is called send rate

TCP Congestion Control (Simon Lam) 9

Effect of Congestion
 W too big for each of many flows -> congestion
 Packet loss -> transmissions on links prior to packet
loss are wasted
 Congestion collapse due too many retransmissions
and too much waste
 October 1986, Internet had its first congestion
collapse
goodput

load
TCP Congestion Control (Simon Lam) 10

5
TCP Window Control

 Receiver flow control

 Avoid overloading receiver
 rwnd: receiver (advertised) window
 Receiver sends rwnd to sender

 Network congestion control

 Sender tries to avoid overloading network
 It infers available network capacity from “loss
i di ti
indications”
”
 cwnd: congestion window

 Sender sets W = min (cwnd, rwnd)

TCP Congestion Control (Simon Lam) 11

Receiver Flow Control

 Receiver advertises rwnd with each packet it
sends
 Size of rwnd indicates available space in
receive buffer
 decreased when data is received from IP layer and
ack’d
 increased when data is consumed by application
process

TCP Congestion Control (Simon Lam) 12

6
Network Congestion Control
 Sender calculates cwnd from indications of
network congestion
 Congestion indications
 timeout (loss)
 dupACK (loss likely)
 queueing delay
 mark (needs ECN)
 TCP algorithms to calculate cwnd
 Tahoe, Reno, Vegas, …
 Link algorithms:
 RED, REM …

TCP Congestion Control (Simon Lam) 13

TCP & AQM

pl((t))

xi(t)

Congestion measures pl(t) for distributed feedback control of xi(t)

 loss and dupACK (DropTail)
 queueing delay (Vegas)
with the help of active queue management (AQM)
 queue length (RED)
 price (REM)
TCP Congestion Control (Simon Lam) 14

7
TCP Congestion Control
 Tahoe (Jacobson 1988)
 Slow Start
 Congestion Avoidance
 Fast
F t Retransmit
R t it
 Reno (Jacobson 1990)
 Fast Recovery
 Its variants: NewReno, SACK
 Vegas (Brakmo & Peterson 1994)
 New Congestion Avoidance
 AQM
 RED
E (Floyd
(Fl d & Jacobson
b 1993)
1 )
• Probabilistic marking or dropping
 REM (Athuraliya & Low 2000)
• Clear buffer, match rate
 Others…

TCP Congestion Control (Simon Lam) 15

Slow Start
 Start with cwnd = 1
 On each successful ACK, increment cwnd
cwnd  cwnd + 1
 Exponential growth of cwnd
each RTT: cwnd  2 x cwnd

 Enter CA when cwnd >= ssthresh

 For initial slow start,, ssthresh is set to a very

y large
g
value (e.g., 65 Kbytes)

Note: for clarity, cwnd, rwnd, and ssthresh are

counted in packets (segments) rather than in bytes

TCP Congestion Control (Simon Lam) 16

8
 Slow Start
sender receiver
cwnd
1
data packet
1 RTT
ACK
2

3
4

5
6
7
8

cwnd  cwnd + 1 (for each ACK)TCP Congestion Control (Simon Lam) 17

Congestion Avoidance
sender receiver  CA starts when
cwnd
1 cwnd  ssthresh
data packet

ACK
 On each successful
2 1 RTT
ACK:
cwnd  cwnd + 1/cwnd
3

 Linear growth of cwnd

4
each RTT:
cwnd  cwnd + 1

TCP Congestion Control (Simon Lam) 18

9
 Packet Loss
 Assumption: loss indicates congestion
 Packet
P k t loss
l detected
d t t db by
 Retransmission timeout (RTO timer)
 Duplicate ACKs (at least 3)

Packets
1 2 3 4 5 6 7

Acknowledgements
1 2 3 3 3 3
TCP Congestion Control (Simon Lam) 19

Fast Retransmit
 A timeout is quite long (> RTT)
 Upon
p receiving
g 3 dupACKs,
p , immediately
y
retransmit without waiting for timeout

 Adjusts ssthresh
ssthresh  max(flightsize/2, 2)
where
h flightsize
fli htsiz is numb
number of
f outstanding
utst ndin ppackets,
ck ts
which may be less than W = min(rwnd, cwnd)

 Enter Slow Start (cwnd = 1)

TCP Congestion Control (Simon Lam) 20

10
TCP Tahoe (Jacobson 1988)

cwnd

time
SS CA
(in RTTs)

SS: Slow Start

CA: Congestion Avoidance

TCP Congestion Control (Simon Lam) 21

Successive Timeouts
 When there is another timeout, double the
timeout value
 Keep doing so for each additional loss-
retransmission
 Exponential backoff up to
max timeout value equal
to 64 times initial timeout
value

Note: red line in figure denotes a loss indication

TCP Congestion Control (Simon Lam) 22

11
Summary: Tahoe
 Basic ideas
 Probe network for spare capacity during SS and
CA and increase send rate
 Drastically reduce rate on congestion indication
 Self-clocking
 Error recovery by retransmission
 Round trip time estimation (to get TO value)
for every ACK {
if (W < ssthresh) then W++ (SS)
else W += 1/W (CA)
}
for every loss indication {
ssthresh = W/2
W = 1
} TCP Congestion Control (Simon Lam) 23

TCP Tahoe (Jacobson 1988)

cwnd

time
SS CA

SS: Slow Start

CA: Congestion Avoidance

TCP Congestion Control (Simon Lam) 24

12
 TCP Reno (Jacobson 1990)

cwnd

time
SS CA

SS: Slow Start

CA: Congestion Avoidance Fast retransmission/fast recovery
TCP Congestion Control (Simon Lam) 25

TCP Reno (another scenario)

TO

cwnd 3 dupACKs

halved
Slow start until cwnd
Initial slow start reaches ssthresh
t

TCP Congestion Control (Simon Lam) 26

13
 Fast recovery (in detail)

 Idea: each dupACK represents a packet

successfully received.
received Therefore,
Therefore no need for
very drastic action
 Enter FR/FR after 3 dupACKs
 Set ssthresh  max(flightsize/2, 2)
 Retransmit lost packet
 Set cwnd  ssthresh + #dupACKs (window inflation)
 Wait till W=min(rwnd
W=min(rwnd, cwnd) is large enough; transmit
new packet(s)
 On non-dup ACK (1 RTT later), set cwnd  ssthresh
(window deflation)
 Enter CA

TCP Congestion Control (Simon Lam) 27

Example: FR/FR

S 1 2 3 4 5 6 7 8 1 9 10 11
time
Exit FR/FR
R 0 0 0 0 0 0 0 8 time

cwnd 8 7 9 11 4
ssthresh 4 4 4 4
 Above scenario: Packet 1 is lost, p
packets 2, 3, and
4 are received;
i d dupACKs
d ACK with
i h seq. no. 0 returned
d
 Fast retransmit
 Retransmit on 3 dupACKs
 Fast recovery
 Inflate window such that new packets 9, 10, and 11 can be
sent while repairing loss
TCP Congestion Control (Simon Lam) 28

14
Summary: Reno
 Basic ideas
 Fast recovery avoids slow start
 dupACKs: fast retransmit + fast recovery
 Timeout: fast retransmit + slow start

dupACKs
congestion
avoidance FR/FR

timeout

slow start retransmit

29
TCP Congestion Control (Simon Lam)

AIMD in steady state

additive increase: multiplicative decrease:
increase cwnd by 1 cut cwnd in half after
MSS every RTT in the 3 dupACKs
absence
bs of
f any lloss
ss
event
congestion
window

24 Kbytes

16 Kbytes

8 Kbytes

time

Long-lived TCP connection

TCP Congestion Control (Simon Lam) 30

15
TCP throughput (send rate)

 We derived the approximate formula

throughput = W packets/sec
RTT
 W changes with the arrival of each
congestion indication
 To calculate
l l (average)
( ) send
d rate, we need
d
the average value of W
Q: W is a function of what parameter?

TCP Congestion Control (Simon Lam) 31

First approximation
M. Mathis, et al., “The Macroscopic Behavior of the TCP Congestion
Avoidance Algorithm,”ACM Computer Communicatons Review, 27(3), 1997.
 No slow-start, no timeout, long-lived TCP
connection
 Independent identically distributed “periods”
 Each packet may be lost with probability p

TCP Congestion Control (Simon Lam) 32

16
 Geometric Distribution
Ave. no. of transmissions to get first loss
 
n   ibi   i (1  p)i1 p
i 1 i 1

 p  i (1  p )i 1
i 1

d  d 
 p 
dp i 1
(1  p )i   p
dp
 (1  p)
i 0
i

d 1 1
 p p 2
dp 1  1  p p
 1/ p

Similarly, ave. no. of transmissions to get first success is

1/(1-p)
TCP Congestion Control (Simon Lam) 33

First approximation (cont.)

 Average number of
8
packets delivered in W
one period (area under 3p
one saw-tooth)
2 2 send rate (in packets/sec)
W  1 W  3 2
      W 3 2
W
 2  2 2  8 no. of packets/period 8
 
 Average number of time per period W 
RTT  
packets
k sent per period
i d  2
(incl. loss at the end) is 1/ p 1 3
 
1/p  2  RTT 2p
RTT  
 Equating the two and  3p 
solving for W, we get
TCP Congestion Control (Simon Lam) 34

17
TCP ACK generation [RFC 1122, RFC 2581]

Event at Receiver TCP Receiver action

Arrival of in-order segment with Delayed ACK. Wait up to 500ms
expected seq #. All data up to for next segment. If no next segment,
expected seq # already ACKed send ACK

Arrival of in-order segment with Immediately send single cumulative

expected seq #. One other ACK, ACKing both in-order segments
segment has ACK pending

Arrival of out
out-of-order
of order segment Immediately send duplicate ACK
ACK,
higher-than-expect seq. # . indicating seq. # of next expected byte
Gap detected

Arrival of segment that Immediate send ACK, provided that

partially or completely fills gap segment starts at lower end of gap

TCP Congestion Control (Simon Lam) 35

Receiver implements Delayed ACKs

 Receiver sends one ACK for every two packets
received -> each saw-tooth is WxRTT wide
-> area under a saw-tooth
h is 3W 2
4
 Send rate is 1 3
RTT 4p
 One ACK for every b packets received -> send rate
i
is

1 3
RTT 2bp
TCP Congestion Control (Simon Lam) 36

18
 Modeling
M d li TCP Throughput:
Th h t A Si Simple
l
Model and its Empirical Validation,
Proc. ACM SIGCOMM, 1998
Jitendra Padhye, Victor Firoiu,
D Towsley,
Don T l andd Jim
Ji Kurose
K

Motivation
 Previous formulas not so accurate when
loss rates are high
 TCP traces show that there are more loss
indications due to timeouts (TO) than due
to triple dupACKs (TD)

TCP Congestion Control (Simon Lam) 38

19
Objectives
 More accurate steady-state throughput
formula as a function of loss rate and RTT
by also accounting for TO behavior of a
TCP connection
 Formula applicable over a wider range of
loss rates
 Explicit statements of assumptions and
approximations
i ti used
d iin d
derivation
i ti of f
throughput formula
 Formula to include the impact of a small
rwnd
TCP Congestion Control (Simon Lam) 39

Many assumptions and

approximations
 A1. TCP sender is saturated, i.e., source
application process always has a packet to
send when send window has space available
 i.e., bulk transfer application
 A2. Slow Start not modeled
 A3. Time to send all packets in a window is
smaller than RTT
 i.e.,transmission rate is not too low

TCP Congestion Control (Simon Lam) 40

20
 A3. Time to send W packets is
less than RTT •ACK reception
marks the end of
current round and
Start of round
beginning of next
time round.
•Approximation: For
b > 1, ACK is not
received
immediately after
one RTT, but it is so
End of round
assumed in the
analysis

space
TCP Congestion Control (Simon Lam) 41

AIMD evolution of Window Size over time

 Each TD period is ended by a TD loss indication.

 TDPi p
period has duration Ai rounds
 A4. Duration of a round (RTT) is independent of
window size
 approximation (poor for a slow line)
 A5. No window inflation in Fast Recovery
 approximation
TCP Congestion Control (Simon Lam) 42

21
 Markov regenerative assumption
 For the i-th TD period, Wi is window size at
the end of the period, Yi is the number of
packets sent in the period
 A6. Assume {Wi} to be a Markov
regenerative process with rewards {Yi}
 Given A6, the steady-state TCP throughput
is
N t E[Yi ] E[Y ]
B  lim Bt  lim  
t  t  t E[ Ai ] E[ A]

TCP Congestion Control (Simon Lam) 43

Consider i-th TD period

when ACK of
last packet is
received

 One ACK after receiving b packets (b = 2 in above

figure) -> linear increase has a slope of 1/b packet per
RTT
 Number of rounds is Xi +1
 i is the first packet lost in i-th TD period
TCP Congestion Control (Simon Lam) 44

22
Loss assumptions
 A7. Losses in different rounds are
independent
 approximation
 A8. Losses within the same round are
correlated as follows: If a packet is lost,
all remaining packets transmitted until the
end of that round are also lost
 approximation
i ti – bursty
b t lloss b behavior
h i b butt only
l
within the same round
 all lost packets in the same round are counted
as a single loss indication when estimating p

TCP Congestion Control (Simon Lam) 45

AIMD throughput derivation (1)

E[ ]  1/ p  seq. no. of first loss
r round trip time
E[r]  RTT
Y no. of p
packets sent
1
E[Y ]  E[ ]  E[W ] 1  1 E[W ] W window size
p X no. of rounds
From Wi 
Wi1 Xi
 , we have A time duration of a period
2 b
b
E[ X ]  E[W ]
2
b
E[ A]  (E[ X ] 1)E[r]  ( E[W ] 1)RTT <- from A4 that round trip
2
times are independent of Wi
1
1 E[W ]
E[Y ] p
send rate B  
E[ A] ( b E[W ] 1)RTT
2 TCP Congestion Control (Simon Lam) 46

23
 AIMD throughput derivation (2)
Another way to
X i / b 1
Wi 1 compute E[Y]
Yi  
k 0 2
 k )b  i
(

XW X X
 i i 1  i ( i  1)  i
2 2 b
X W
 i (Wi  i 1  1)  i
2 2
W
Let E[  ] be E[ ] and we have
2 <- A9. Assume that
E[ X ] E[W ] {Xi} and {Wi} are
E[Y ]  ( E[W ]   1)  E[  ]
2 2 mutually
bE[W ] E[W ] W independent i.i.d.
 ( E[W ]   1)  E[ ]
4 2 2 sequences of
random variables
TCP Congestion Control (Simon Lam) 47

AIMD throughput (3)

 Equate the two previous formulas for E[Y]. Solve the quadratic
equation with E[W] as the only unknown
1
E [Y ]   1  E [W ]
p
b E [W ] E [W ] W
 ( E [W ]   1)  E [ ]
4 2 2

2b 8(1  p ) 2b 2

E [W ]   ( )
3b 3b p 3b

1 p
 E [W ]
p
se n d ra te B( p) 
E [ A]
TCP Congestion Control (Simon Lam) 48

24
AIMD throughput (4)
To get a simple formula, collect terms that are o(1/sqrt(p))

8
E[W ]   o(1/ p )
3bp
b 2b
E[ X ]  E[W ]   o(1/ p )
2 3p
1/ p  o(1/ p) 1 3
send rate B( p)    o(1/ p )
 2b  RTT 2bp
RTT   o(1/ p ) 
 3p 

TCP Congestion Control (Simon Lam) 49

AIMD with TO

 Let ni denote the number of TD periods within a

cycle ending in i-th
th TO period, Ri denote no. of
retransmissions in i-th TO period
 A10. {ni } form an i.i.d. sequence, independent of
{Yij} and {Aij}

TCP Congestion Control (Simon Lam) 50

25
 Throughput of AIMD with TO (1)
E[ M ]  E[n]E[Y ]  E[ R ] Assumption of
Markov
E[ S ]  E[n]E[ A]  E[ Z TO ]
regenerative
E[ M ] E[n]E[Y ]  E[ R] process again.
send rate B  
E[ S ] E[n]E[ A]  E[ Z TO ]
E[Y ]  Q  E[ R]
B
E[ A]  Q  E[ Z TO ]
1
where Q 
E[n] <- Probability
P b bilit ththatt a
1 given loss
E[ R]  indication is a TO
1 p
with Q and E[ Z TO ] to be determined
TCP Congestion Control (Simon Lam) 51

Approximate solution for Q

A given loss indication is a TO is the union of two

events Two or less acked packets in penultimate
round or two or less acked packets in final round
TCP Congestion Control (Simon Lam) 52

26
 Approximate solution for Q (cont.)
<- penultimate round of w
(1  p ) k p packets, first k packets
A( w, k ) 
1  (1  p ) w ack’d given there is a loss

C (k , m )  (1  p ) m p, m  k 1 <- for last round, k packets

sent, m packets ack’d in
C (k , m )  (1  p ) m , mk sequence

 ( w)  1
Q if w  3 <- at most 2 dupACKs
p
2 w 2 <- probability of fewer than 3
  A( w, k )   A( w, k )  C ( k , m) packets sent successfully
k 0 k 3 m 0
in penultimate round or
if w  4 less than 3 acks in last
round
TCP Congestion Control (Simon Lam) 53

Approximate solution for Q (cont.)

 ( w)]
 Q is E[Q
But we don’t know the probability distribution of Wi
 Approximation  3 3bp
Q  Q ( E[W ])  min(1, )  min(1,3 )
E[W ] 8
TCP Congestion Control (Simon Lam) 54

27
 Throughput of AIMD with TO (2)
P[ R  k ]  p k 1 (1  p ) for k  1, 2,...
<- duration of k
Lk  (2  1)TO
k
for k  6 TOs in a row
 (63  64(k  6))TO for k  7
1  p  2 p 2  4 p 3  8 p 4  16 p 5  32 p 6
E[ Z TO ]  TO
1 p

 TO
f ( p)
 T0 (1  32 p 2 ) <- approximation
1 p
E[Y ]  Q  E[ R ]
sendd rate B ( p ) 
E[ A]  Q  E[ Z TO ]
1 p  ( E[W ]) 1
 E[W ]  Q
p 1 p
B( p) 
RTT ( E[ X ]  1)  Q  ( E[W ])T f ( p )
1  p TCP Congestion Control (Simon Lam)
O
55

Throughput of AIMD with TO (3)

1 p  ( E[W ]) 1
 E[W ]  Q <- Eq. (27)
p 1 p
B( p)  more accurate
 ( E[W ])T f ( p)
RTT ( E[ X ]  1)  Q version of
1 p
O
throughput
1/ p formula

 2b   3bp 
RTT    min 1,3  (1  32 p )T0
2

 3 p   8 
1
 <-Eq. (29)
 2bp   3bp  most well-
RTT    min 1,3  p(1  32 p )T0
2

 3   8  known version
of throughput
formula
TCP Congestion Control (Simon Lam) 56

28
 Impact of receiver’s rwnd limitation

Compute E[W ]. If E[W ]  Wmax , use Eq. (27): Full model Eq. (31)
1 p  ( E[W ]) 1
 E[W ]  Q
p 1 p
B( p)  if E[W] <Wmax,
 ( E[W ])T f ( p)
RTT ( E[ X ]  1)  Q
1 p
O

1 p  (W ) 1
 Wmax  Q otherwise, use Wmax for
1 p
max
p E[W] and recompute
B( p) 
b 1 p  f ( p) E[X]
RTT ( Wmax   2)  Q(Wmax )TO
8 pWmax 1 p (derivation omitted)
TCP Congestion Control (Simon Lam) 57

Impact of receiver’s rwnd

limitation—approximate model
Use the well-known Eq. (29) from before,

Wmax 1
B( p )  min( , )
RTT  2bp   3bp 
RTT    min 1,3  p (1  32 p )T0
2

 3   8 

which is referred to as Eq. (32)

TCP Congestion Control (Simon Lam) 58

29
Summary data from traces (1 hour)
 Saturated TCP
sender
 p computed
from dividing
total no. of loss
indications by
total number of
packets sent
 RTT and TO
values are
averaged over
entire 1-hour
trace

TCP Congestion Control (Simon Lam) 59

Summary data from 100s traces

 Each row represents results of 100 traces each of

100 seconds in duration for same S-D pair
 Totals are cumulative over 100 traces
 RTT and TO are average values over 100 traces for
same S-D pair TCP Congestion Control (Simon Lam) 60

30
Experimental comparison (1)

 Each point represents number of packets in 100s interval of trace

 T0 ~ single TO, T1 ~ at least 1 double TO in trace, etc.
 “TD Only” is analytic model by Mathis et al.
 Note: Wmax is only 6 in Figure 7 TCP Congestion Control (Simon Lam) 61

Experimental comparison (2)

Wmax = 33 Wmax=44
TCP Congestion Control (Simon Lam) 62

31
Experimental comparison (3)

Wmax=8 Wmax=48
TCP Congestion Control (Simon Lam) 63

Accuracy of approximate model

Figure 18: manic to spiff, with predictions by both full and

approximate models (Wmax=32)
TCP Congestion Control (Simon Lam) 64

32
Average errors

N predicted  N observed
 N observed
ave. error  observations

no. of observations TCP Congestion Control (Simon Lam) 65

Conclusions
 A much more rigorous analysis than the one by
Mathis et al.
 Numerous assumptions and approximations used
but (almost) all of them are explicitly stated
 Large amount of experimental measurements on
the Internet to validate accuracy of the full model
(less for the approximate model)
 Throughput formula accounts for loss indications
due to TO as well as rwnd restriction
 Using
U i theh formula
f l requires
i accurate measurements of
f
loss rate and RTT values (which could be tricky)
 For TCP Reno and drop-tail router
 Accuracy (like beauty) is in the eye of the
beholder. What do you think?
TCP Congestion Control (Simon Lam) 66

33
TCP Throughput limited by loss rate
 TCP average throughput (approximate) in
terms of loss rate, L:
1 22  MSS
1.22
RTT p
 Example: 1500-byte segments, 100ms RTT,
to get 10 Gbps throughput, loss rate needs
to be very low
p = 2x10
2 10-10
10

 New version of TCP needed for connections

with high-delay bandwidth product
 addressed in paper by Katabi’s et al
TCP Congestion Control (Simon Lam) 67

The End

TCP Congestion Control (Simon Lam) 68

34