37
Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements • Appropriate statistics Connecting traffic to application behavior “optimal” web layout HT files HT traffic Traff ic Verbal Data/ stat Mod/sim Analysi s Synthes is

Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Embed Size (px)

Citation preview

Page 1: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Traffic (1993-2000)

• Heavy tails (HT) in net traffic???

• Careful measurements• Appropriate statistics• Connecting traffic to

application behavior• “optimal” web layout

HT files

HT traffic

Traffic

Verbal

Data/stat

Mod/sim

Analysis

Synthesis

Page 2: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Is streamed out on the net.

Creating fractal Gaussian internet traffic (Willinger,…)

2

3 H

Heavy tailed files

time

log(file size)

> 1.0

log(

> s

ize)

p s-

Page 3: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Traffic (1993)

• Traffic is “bursty”?Traffic

Verbal

Page 4: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Traffic (1993-2000)

• Bursty???• Careful measurements• Appropriate statistics

Traffic

Verbal

Data/stat

Why?

Page 5: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Heavy tailed files

time

Long space

Becomes long time

Why?

Traffic

Verbal

Data/stat

Mod/sim

Page 6: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Heavy tailed files

time

log(file size)

> 1.0

log(

> s

ize)

p s-

2

3 H

Traffic

Verbal

Data/stat

Mod/sim

Analysis

Page 7: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Heavy tailed files

time

log(file size)

> 1.0

log(

> s

ize)

p s-

What?

Page 8: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency

Decimated dataLog (base 10)

Forest fires1000 km2

(Malamud)

WWW filesMbytes

(Crovella)

Data compression

(Huffman)

Cumulative

Page 9: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

log( ( ))P X x

log( )x

cx Probability that a file is bigger than x.

1cx

Probability that a packet is in a file bigger than x.

Page 10: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

FrequencyFires

Web filesCodewords

Cumulative

Log (base 10)

-1/2

-1

Page 11: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency Forest fires1000 km2

WWW filesMbytes

Data compression

Cumulative

-1/2

-1

exponential

Page 12: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency Forest fires1000 km2

WWW filesMbytes

Data compression

Cumulative

exponential

All events are close in size.

Page 13: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency Forest fires1000 km2

WWW filesMbytes

Data compression

Cumulative

-1/2

-1

Most events are small

But the large events are huge

Page 14: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

FF

WWWDC

Data + Model/Theory

Page 15: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency

Decimated dataLog (base 10)

WWW filesMbytes

(Crovella)

Cumulative Most files are small

(mice)

Most packets are in large files (elephants)

Page 16: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

NetworkNetwork

Sources

Mice

Elephants

Router queues

Delay sensitive

Bandwidth sensitive

Unfortunate interaction of files with congestion

control

Page 17: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Heavy tailed files

time

log(file size)

> 1.0

log(

> s

ize)

p s-

Why?

Page 18: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

Size of events

Frequency

WWW filesMbytes

Data compression

Cumulative

exponential

All events are close in size.

Page 19: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Source coding for data compression

Based on frequencies of source word occurrences,

Select code words.

To minimize message length.

Page 20: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

0 1 2-1

0

1

2

3

4

5

6

DC

Data

Avg. length =

log( )

i i

i i

p l

p p

How well does the model predict the data?

length log(

xp( )

)

ei i

i i

l p

p cl

Page 21: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

0 1 2-1

0

1

2

3

4

5

6

DC

Data + Model

How well does the model predict the data?

Not surprising, because the file was compressed using

Shannon theory.

Small discrepancy due to integer lengths.

length log(

xp( )

)

ei i

i i

l p

p cl

Avg. length =

log( )

i i

i i

p l

p p

Page 22: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Generalized “coding” problems

• Minimize avg file transfer• No feedback• Discrete (0-d) topology

• Minimize avg file transfer• Feedback• 1-d topology

Web

Data compression

Page 23: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

document

split into N files to minimize download time

A toy website model(= 1-d grid HOT design)

Traffic

Verbal

Data/stat

Mod/sim

Analysis

Synthesis

Page 24: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Probability of user access

Wasteful

Hard to navigate.

Page 25: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Wasteful

Hard to navigate.

Just right

Page 26: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

More complete website models

(Zhu, Yu)

• Detailed models – user behavior – content and hyperlinks

• Necessary for real web layout optimization• Statistics consistent with simpler models• Improved protocol design (TCP)• Commercial implications still unclear

Page 27: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Traffic (1993-2000)

• Heavy tails (HT) in net traffic???

• Careful measurements• Appropriate statistics• Connecting traffic to

application behavior• “optimal” web layout

HT files

HT traffic

Traffic

Verbal

Data/stat

Mod/sim

Analysis

Synthesis

Page 28: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

WWWDC

Data

Page 29: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

WWWDC

Data + Model/Theory

Page 30: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

WWW

Data + Model/Theory

Are individual websites distributed like this?

Roughly, yes.

Page 31: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

-6 -5 -4 -3 -2 -1 0 1 2-1

0

1

2

3

4

5

6

WWWDC

Data + Model/Theory

How has the data changed since 1995?

Page 32: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Traffic (1993-2000)

Traffic Topology Layering C&D

Verbal

Data/stat

Mod/sim

Analysis

Synthesis

Page 33: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

Theory and the Internet

Traffic Topology C&D Layering

Verbal

Data/stat

Mod/sim

Analysis

Synthesis

Page 34: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

NetworkNetwork

Sources

Mice

Elephants

Router queues

Page 35: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

NetworkNetwork

Sources

Mice

Elephants

Router queues

Delay sensitive

Bandwidth sensitive

Unfortunate interaction of files with congestion

control

Page 36: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

NetworkNetwork

Sources

Mice

Elephants

Router queues

Delay sensitive

Bandwidth sensitive

Better Control

Fortunate interaction of files with improved congestion control

Page 37: Traffic (1993-2000) Heavy tails (HT) in net traffic??? Careful measurements Appropriate statistics Connecting traffic to application behavior “optimal”

High variability in context

More high variability• Heterogeneity• Human behavior• Actuating

Today: • Simplify/broaden • Look back/sideways

Extend• Optimization• Layer/distribute• Dynamics/control

Develop• Delays• Actuation