BitTorrent Experiments on Testbeds: A Study of the Impact of Network - - PowerPoint PPT Presentation

bittorrent experiments on testbeds a study of the impact
SMART_READER_LITE
LIVE PREVIEW

BitTorrent Experiments on Testbeds: A Study of the Impact of Network - - PowerPoint PPT Presentation

Title Introduction Methodology Experimental Results Conclusion Backup BitTorrent Experiments on Testbeds: A Study of the Impact of Network Latencies Ashwin Rao , Arnaud Legout, and Walid Dabbous INRIA, Projet Plan` ete


slide-1
SLIDE 1

Title Introduction Methodology Experimental Results Conclusion Backup

BitTorrent Experiments on Testbeds: A Study of the Impact of Network Latencies

Ashwin Rao, Arnaud Legout, and Walid Dabbous

INRIA, Projet Plan` ete (ashwin.rao,arnaud.legout,walid.dabbous)@inria.fr

1/17

slide-2
SLIDE 2

Title Introduction Methodology Experimental Results Conclusion Backup

Outline

Introduction Methodology Experimental Results Homogeneous Latency Heterogeneous Latency Conclusion

2/17

slide-3
SLIDE 3

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-4
SLIDE 4

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-5
SLIDE 5

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-6
SLIDE 6

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-7
SLIDE 7

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-8
SLIDE 8

Title Introduction Methodology Experimental Results Conclusion Backup

Overview of BitTorrent

3/17

slide-9
SLIDE 9

Title Introduction Methodology Experimental Results Conclusion Backup

Evaluation of BitTorrent Performance

Planetlab sites [planete-lab.org] Grid5000 sites [grid5000.fr]

  • Grid5000 and PlaneteLab Testbeds

4/17

slide-10
SLIDE 10

Title Introduction Methodology Experimental Results Conclusion Backup

Evaluation of BitTorrent Performance

Planetlab sites [planete-lab.org] Grid5000 sites [grid5000.fr]

  • Grid5000 and PlaneteLab Testbeds
  • Absence of network latency between
  • Instances of application running on same machine
  • Machines in the same LAN

4/17

slide-11
SLIDE 11

Title Introduction Methodology Experimental Results Conclusion Backup

Evaluation of BitTorrent Performance

Planetlab sites [planete-lab.org] Grid5000 sites [grid5000.fr]

  • Grid5000 and PlaneteLab Testbeds
  • Absence of network latency between
  • Instances of application running on same machine
  • Machines in the same LAN

Does network latency affect the outcome of BitTorrent experiments performed on testbeds

4/17

slide-12
SLIDE 12

Title Introduction Methodology Experimental Results Conclusion Backup

Can Latency Impact BitTorrent Performance?

TCP Ramp-up and Impact of Upload Rates

5/17

slide-13
SLIDE 13

Title Introduction Methodology Experimental Results Conclusion Backup

Can Latency Impact BitTorrent Performance?

TCP Ramp-up and Impact of Upload Rates Delays in receiving BitTorrent control messages

5/17

slide-14
SLIDE 14

Title Introduction Methodology Experimental Results Conclusion Backup

Can Latency Impact BitTorrent Performance?

TCP Ramp-up and Impact of Upload Rates Delays in receiving BitTorrent control messages Can testbeds such as Grid5000 be used for experimental evaluation of BitTorrent?

5/17

slide-15
SLIDE 15

Title Introduction Methodology Experimental Results Conclusion Backup

Outline

Introduction Methodology Experimental Results Homogeneous Latency Heterogeneous Latency Conclusion

6/17

slide-16
SLIDE 16

Title Introduction Methodology Experimental Results Conclusion Backup

Testbed Setup

4 Machines of Grid5000 Experimental Testbed

7/17

slide-17
SLIDE 17

Title Introduction Methodology Experimental Results Conclusion Backup

Testbed Setup

Machines capable of running 100 instances of BitTorrent Client

7/17

slide-18
SLIDE 18

Title Introduction Methodology Experimental Results Conclusion Backup

Testbed Setup

Peers on same machine communicate using Loopback device

7/17

slide-19
SLIDE 19

Title Introduction Methodology Experimental Results Conclusion Backup

Testbed Setup

Peers on different machines communicate using Ethernet device

7/17

slide-20
SLIDE 20

Title Introduction Methodology Experimental Results Conclusion Backup

Testbed Setup

Tracker and Seed placed on same machine

7/17

slide-21
SLIDE 21

Title Introduction Methodology Experimental Results Conclusion Backup

Experiment Setup

  • Experiment Parameters
  • 50 MB file
  • 1 Tracker, 1 Seed, and 300 Leechers
  • Upload rates - 10 KiB/s to 100 KiB/s
  • Emulate RTT - 1000 ms
  • Metric
  • Download completion time

8/17

slide-22
SLIDE 22

Title Introduction Methodology Experimental Results Conclusion Backup

Outline

Introduction Methodology Experimental Results Homogeneous Latency Heterogeneous Latency Conclusion

9/17

slide-23
SLIDE 23

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Setup

Same delay on Loopback and Ethernet Device

10/17

slide-24
SLIDE 24

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Download Completion Time vs RTT

11/17

slide-25
SLIDE 25

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

1000 ms RTT does not increase Average Download Completion Time by more than 15%

11/17

slide-26
SLIDE 26

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Download Completion Time not a monotonously increasing function of RTT

11/17

slide-27
SLIDE 27

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Download Completion Time not a monotonously increasing function of RTT

11/17

slide-28
SLIDE 28

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Impact of latency not observed when seed is fast

11/17

slide-29
SLIDE 29

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Marginal impact of RTT

11/17

slide-30
SLIDE 30

Title Introduction Methodology Experimental Results Conclusion Backup

Homogeneous Latency - Experiment Results

Marginal impact of RTT

11/17

slide-31
SLIDE 31

Title Introduction Methodology Experimental Results Conclusion Backup

Heterogeneous Latency

Each machine used to abstract an AS

12/17

slide-32
SLIDE 32

Title Introduction Methodology Experimental Results Conclusion Backup

Heterogeneous Latency

Different latency on loopback and ethernet device

12/17

slide-33
SLIDE 33

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Small RTT Between Peers

Upload Rate limited to 20 kB/s. RTT between a pair of leechers. RTT between any two peers is less than 100 ms

13/17

slide-34
SLIDE 34

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Small RTT Between Peers

Upload Rate limited to 20 kB/s. RTT between a pair of leechers. RTT between any two peers is less than 100 ms Marginal Impact of Latency

13/17

slide-35
SLIDE 35

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Small RTT Between Peers

Upload Rate limited to 20 kB/s. RTT between a pair of leechers. RTT between any two peers is less than 100 ms Marginal Impact of Latency

13/17

slide-36
SLIDE 36

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Large RTT Between Some Peers

Upload Rate limited to 20 kB/s. RTT between a pair of leechers. RTT between some of the peers is greater than 200 ms

14/17

slide-37
SLIDE 37

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Large RTT Between Some Peers

Upload Rate limited to 20 kB/s. RTT between a pair of leechers. RTT between some of the peers is greater than 200 ms Marginal Impact of Latency

14/17

slide-38
SLIDE 38

Title Introduction Methodology Experimental Results Conclusion Backup

Outline

Introduction Methodology Experimental Results Homogeneous Latency Heterogeneous Latency Conclusion

15/17

slide-39
SLIDE 39

Title Introduction Methodology Experimental Results Conclusion Backup

Conclusion

  • Latency has a marginal impact on the outcome of

BitTorrent experiments

16/17

slide-40
SLIDE 40

Title Introduction Methodology Experimental Results Conclusion Backup

Conclusion

  • Latency has a marginal impact on the outcome of

BitTorrent experiments BitTorrent experiments can be performed on testbeds without explicitly emulating latency.

16/17

slide-41
SLIDE 41

Title Introduction Methodology Experimental Results Conclusion Backup

BitTorrent Experiments on Testbeds: A Study of the Impact of Network Latencies

Ashwin Rao (ashwin.rao@inria.fr)

17/17

slide-42
SLIDE 42

Title Introduction Methodology Experimental Results Conclusion Backup

TCP Segmentation Offloading

0.2 0.4 0.6 0.8 1 1 10 100 1000 10000 TCP Payload Length (bytes) TSO Disabled 0 ms 100 ms 400 ms 0.2 0.4 0.6 0.8 1 1 10 100 1000 10000 CDF WIDE backbone WIDE backbone 0.2 0.4 0.6 0.8 1 1 10 100 1000 10000 TSO Enabled 0 ms 100 ms 400 ms

Figure: Impact of Segmentation Offloading

−/16

slide-43
SLIDE 43

Title Introduction Methodology Experimental Results Conclusion Backup

Number of Nodes on a Machine

100 200 300 400 500 600 700 800 900 1000 1100 25 50 75 100

RTT estimate of TCP (ms)

Number of leechers running on a given machine

Minimum RTT 0 ms Minimum RTT 400 ms Minimum RTT 1000 ms

Figure: Impact of Number of Nodes on a Machine

Machines support up to 100 instances of a leecher

−/16

slide-44
SLIDE 44

Title Introduction Methodology Experimental Results Conclusion Backup

Send Call Inter-arrival Time

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 <20 40 70 90 120 190 275 420 550 >1000

CDF

Inter arrival time of data messages

  • n a given connection (ms)

10 kB/s 20 kB/s 50 kB/s 100 kB/s 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 <20 40 70 90 120 190 275 420 550 >1000

CDF

Inter arrival time of data messages

  • n a given connection (ms)

10 kB/s 20 kB/s 50 kB/s 100 kB/s

−/16

slide-45
SLIDE 45

Title Introduction Methodology Experimental Results Conclusion Backup

RTT between pair of peers

AS1 AS2 AS3 AS4 AS5 AS1 8 ms 40 ms 60 ms 210 ms 210 ms AS2 40 ms 20 ms 80 ms 230 ms 230 ms AS3 60 ms 80 ms 40 ms 250 ms 250 ms AS4 210 ms 230 ms 250 ms 100 ms 400 ms AS5 210 ms 230 ms 250 ms 400 ms 200 ms

Table: RTT between a pair of leechers.

AS1 AS2 AS3 AS4 AS5 AS′

1

20 ms 40 ms 60 ms 210 ms 210 ms AS′

2

40 ms 60 ms 80 ms 230 ms 230 ms AS′

3

60 ms 80 ms 100 ms 250 ms 250 ms AS′

4

210 ms 230 ms 250 ms 400 ms 400 ms AS′

5

210 ms 230 ms 250 ms 400 ms 400 ms

Table: RTT between the initial seed and the leechers.

−/16

slide-46
SLIDE 46

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Large RTT Between Some Peers

Figure: Upload rate limited to 50 KiB/s. RTT between some peers more than 100 ms.

RTT of even 400 ms has marginal impact on download completion time

AS1 AS3 AS5 AS1 8 ms 60 ms 210 ms AS3 60 ms 40 ms 250 ms AS5 210 ms 250 ms 200 ms AS′

1

20 ms 60 ms 210 ms AS′

5

210 ms 250 ms 400 ms

Table: RTT between a pair of leechers.

−/16

slide-47
SLIDE 47

Title Introduction Methodology Experimental Results Conclusion Backup

Results - Large RTT Between Some Peers

Figure: Upload rate limited to 50 KiB/s. RTT between some peers more than 100 ms.

RTT of even 400 ms has marginal impact on download completion time

AS1 AS3 AS5 AS1 8 ms 60 ms 210 ms AS3 60 ms 40 ms 250 ms AS5 210 ms 250 ms 200 ms AS′

1

20 ms 60 ms 210 ms AS′

5

210 ms 250 ms 400 ms

Table: RTT between a pair of leechers.

−/16