performance evaluation of tcp over multiple paths in fixed robust routing

21
Performance Evaluation of TCP over Multiple Paths in Fixed Robust Routing Wenjie Chen, Yukinobu Fukushima, Takashi Matsumura, Yuichi Nishida, and Tokumi Yokohira The Graduate School of Natural Science and Technology, Okayama University, Japan 1 CQR 2011

Upload: otis

Post on 01-Feb-2016

45 views

Category:

Documents


0 download

DESCRIPTION

Performance Evaluation of TCP over Multiple Paths in Fixed Robust Routing. Wenjie Chen, Yukinobu Fukushima , Takashi Matsumura, Yuichi Nishida, and Tokumi Yokohira The Graduate School of Natural Science and Technology, Okayama University, Japan. Background. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Performance Evaluation of TCP over Multiple Paths

in Fixed Robust Routing

Wenjie Chen, Yukinobu Fukushima, Takashi Matsumura, Yuichi Nishida, and Tokumi Yokohira

The Graduate School of Natural Science and Technology,Okayama University, Japan

1CQR 2011

Page 2: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

BackgroundBackground

Penetration of bandwidth-consuming applications(e.g., P2P file sharing and video streaming)

Traffic patterns in ISP networks become variableNeed for ISP networks to accommodate those

variable traffic patterns

Routing for variable traffic patterns Dynamic routing

Increases operational complexity Can lead to route instability

Fixed robust routing [1, 3] Low operational complexity No route instability (static routing)

CQR 2011 2

[1] M. Kodialam, T. V. Lakshman, and S. Sengupta, “Maximum throughput routing of traffic in the hose model,” in Proceedings of IEEE INFOCOM2006, pp. 1–11, Apr. 2006. [3] V. Tabatabaee, A. Kashyap, B. Bhattacharjee, R. J. La, and M. A. Shayman, “Robust routing with unknown traffic matrices,” in Proceedings of IEEE INFOCOM 2007, pp. 2436–2440, May 2007.

Page 3: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Fixed Robust RoutingFixed Robust Routing

Tries to achieve the best worst-case performance (e.g., maximum link load), given variable traffic patterns Traffic patterns are assumed to vary within the region

specified by some traffic variation models (e.g., hose model)

Performs multipath routing Traffic of every source-destination pair is routed on multiple

paths

  Multipath routing causes out-of-order packet arrivals

TCP performance may be degraded

CQR 2011 3

Page 4: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Research ObjectiveResearch Objective

Investigation of TCP performance over general fixed robust routing

Proposal of fixed robust routing algorithm that tries to improve TCP performance in addition to decreasing maximum link load

CQR 2011 4

Page 5: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Formulation of Fixed Robust Routing Problem [3]Formulation of Fixed Robust Routing Problem [3]

CQR 2011 5

Input

Output

: Candidate paths of every (i, j) pair

: Set of traffic matrices that follow hose and pipe traffic model

: Maximum link load

: Fraction of traffic of the corresponding (i, j) pair routed on path p

Linear semi-infinite programming problem(convertible to polynomial size linear programming problem [3])

: Set of all links in the network

: capacity of link l

Path 1

Path 2

Path 3

[3] V. Tabatabaee, A. Kashyap, B. Bhattacharjee, R. J. La, and M. A. Shayman, “Robust routing with unknown traffic matrices,” in Proceedings of IEEE INFOCOM 2007, pp. 2436–2440, May 2007.

subject to

Node i Node j

: Set of paths routed on link l

Page 6: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Performance Degradation of TCP over Fixed Performance Degradation of TCP over Fixed Robust RoutingRobust Routing

CQR 2011 6

Time

Reception of threeduplicated Acks

Time

Source Destination

1

2

3

4

1

2

3

4

Packets on shorter path overtakepreceding packets on longer path

Out-of-order packet arrivals at destination host

Source host receives three duplicatedAcks and decreases its congestionwindow size

TCP throughput is degraded

Shorter Path

Longer Path

4 3

2

1SourceDestination

Page 7: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Evaluation of TCP Performance over Fixed Evaluation of TCP Performance over Fixed Robust Routing: Simulation modelRobust Routing: Simulation model

Two kinds of path between R1 and R2 L (Long path): 2.0 + d [ms] S (Short path): 2.0 [ms]

Combination of paths: SLLL, SSLL, SSSL One TCP connection for every end-host pair (Si , Di)

Si ’s data transmission rate: 20 [Mbps]

CQR 2011 7

S1

S2

S3

S4

S5

R1 R2

D1

D2

D3

D4

D5

Bandwidth: 50 [Mbps]Propagation delay: 0.2 [ms]

Bandwidth: 100 [Mbps]Propagation delay: 2.0 + d [ms] for L 2.0 [ms] for S

Page 8: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Evaluation of TCP Performance over Fixed Evaluation of TCP Performance over Fixed Robust Routing: ResultRobust Routing: Result Larger delay difference

more candidates for overtaking packet Higher ratio of shorter path

higher probability of three out-of-order packet arrivals

CQR 2011 8

d (delay difference between path L and path S) [ms]

0

20

40

60

80

100

0 0.4 0.8 1.2 1.6 2 2.4 2.8

Tota

l th

rou

gh

pu

t [M

bp

s]

d [ms]0.4 0.8 1.2 1.6 2.0 2.4 2.80

0

20

40

60

80

100

Tot

al t

hro

ughp

ut [

Mbp

s]SLLL

SSLL

SSSL

Lower TCP throughput

Page 9: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Proposal of Fixed Robust Routing Taking Account of TCP Proposal of Fixed Robust Routing Taking Account of TCP Performance (1/2): Basic StrategyPerformance (1/2): Basic Strategy

CQR 2011 9

Input

Output

: Maximum link loadsubject to

Linear semi-infinite programming problem

Our proposed fixed robust routing selects such candidate paths ( ) that avoid TCP performance degradation as much as possible

: Set of all links in the network

: capacity of link l: Set of paths routed on link l

: Set of traffic matrices that follow hose and pipe traffic model

: Candidate paths of every (i, j) pair

: Fraction of traffic of the corresponding (i, j) pair routed on path p

Page 10: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Proposal of Fixed Robust Routing Taking Account of TCP Proposal of Fixed Robust Routing Taking Account of TCP Performance (2/2): AlgorithmPerformance (2/2): Algorithm

10

Step. 1 Selection of candidate paths of every source-destination pair Step. 1.1 We select K shortest hop paths Step. 1.2 From the K paths, we select M paths with the minimum delay difference between the shortest and the longest delay pathsStep. 2. We solve the formulated problem and obtain maximum link load (t) and fraction (xp) of traffic routed on every path. When solving the problem, we bound fraction of traffic routed on the shortest delay path by α

Path 1, 15ms

Path 2, 8ms

Path 3, 3ms

Path 4, 14ms

Path 5, 10ms

Node i Node j

MDD-LF (Minimum Delay Difference with Limited Fraction)

Page 11: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Simulation ModelSimulation Model

One TCP connection for every node-pair (Ri , Rj) Source host’s data transmission rate: 10 [Mbps] Parameter settings in MDD-LF

K = 5 M = 2 α = 0.25

Comparison: k-shortest A straightforward fixed robust routing algorithm that selects M (= 2) shortest hop paths as candidate paths for every node-pair

CQR 2011 11

R4[2%]

4.7ms

2.8ms

7.0ms

3.5ms 2.8ms 3.5ms

3.5ms

3.5ms

3.5ms3.5ms

8.4ms

8.4ms

4.9ms

5.6ms

11.2ms

2.8ms

9.1ms

0.7ms

1.4ms

1.4ms

1.4msLink bandwidth: 1 [Gbps]

Page 12: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

0

0.2

0.4

0.6

0.8

1

k-shortest MDD-LF

Ma

xim

um

lin

k lo

ad

0

2

4

6

8

10

k-shortest MDD-LF

Th

rou

gh

pu

t [M

bp

s]Evaluation ResultsEvaluation Results

Compared to k-shortest, MDD-LF: 27% higher throughput

Candidate path selection policy of MDD-LF is effective for improving TCP throughput

CQR 2011 12

0

2

4

6

8

10

Ave

rag

e T

hrou

ghpu

t [M

bps]

k-shortest MDD-LF k-shortest MDD-LF0

0.2

0.4

0.6

0.8

1

Max

imum

link

load

Compared to k-shortest, MDD-LF: 2.3 times higher load MDD-LF tends to select longer hop paths than k-shortest

Page 13: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Conclusions and Future WorkConclusions and Future Work

Conclusions Investigation of TCP throughput over fixed robust routing

Larger delay difference Higher ratio of shorter path

Proposal of fixed robust routing algorithm that tries to improve TCP throughput

MDD-LF: 27% higher throughput but 2.3 times higher load

Future work Performance evaluation of our proposed algorithm in detail Modification of our proposed algorithm

Selection of link-disjoint paths as candidate paths

CQR 2011 13

Lower TCP throughput

Page 14: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Number of Candidates for Overtaking packetsNumber of Candidates for Overtaking packets

CQR 2011 14

d (delay difference between path L and path S) [ms]

0

20

40

60

80

100

0 0.4 0.8 1.2 1.6 2 2.4 2.8

Tota

l th

rou

gh

pu

t [M

bp

s]

d [ms]0.4 0.8 1.2 1.6 2.0 2.4 2.80

0

20

40

60

80

100

Tot

al t

hro

ughp

ut [

Mbp

s] SLLL

SSLL

SSSL

# of candidates for overtaking packets

0 1 2 3 4 5 6

TimeTime

Source Destination

1

2

3

4

1

2

3

4

d = 1.0

0.4

Average packet transmission interval

Page 15: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Evaluation of TCP Performance over Fixed Evaluation of TCP Performance over Fixed Robust Routing: ResultRobust Routing: Result Larger delay difference

more candidates for overtaking packet Higher ratio of shorter path

higher probability of three out-of-order packet arrivals

SLLL: 0.012 SSLL: 0.063 SSSL: 0.11

CQR 2011 15

d (delay difference between path L and path S) [ms]

0

20

40

60

80

100

0 0.4 0.8 1.2 1.6 2 2.4 2.8

Tota

l th

rou

gh

pu

t [M

bp

s]

d [ms]0.4 0.8 1.2 1.6 2.0 2.4 2.80

0

20

40

60

80

100

Tot

al t

hro

ughp

ut [

Mbp

s]SLLL

SSLL

SSSL

Lower TCP throughput

Page 16: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Traffic Variation Models Assumed in Fixed Traffic Variation Models Assumed in Fixed Robust RoutingRobust Routing

Hose traffic model

Pipe traffic model

CQR 2011 16

T =

t11 t12 t1n ・・・t21 t22 ・・・ t2n

t21 t22 ・・・ t2n

・・・

・・・

・・・・・・

: Upper bound on traffic volume that enters the network at node i (e.g., bandwidth of external ingress link of node i)

: Upper bound on traffic volume that leaves the network at node j (e.g., bandwidth of external egress link of node j)

T =

t11 t12 t1n ・・・t21 t22 ・・・ t2n

t21 t22 ・・・ t2n

・・・

・・・

・・・・・・

: Upper bound on traffic volume from node i to node j (The value is determined based on traffic histories or service level agreement)

Page 17: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Evaluation ResultsEvaluation Results

Compared to k-shortest, MDD: 22% higher throughput MDD-LF: 27% higher throughput

candidate path selection policy of MDD and MDD-LD are effective for improving TCP throughput

CQR 2011 17

0

2

4

6

8

10

k-shortest MDD MDD-LF

Th

rou

gh

pu

t [M

bp

s]

0

0.2

0.4

0.6

0.8

1

k-shortest MDD MDD-LF

Ma

xim

um

lin

k lo

ad

0

2

4

6

8

10

Ave

rag

e T

hrou

ghpu

t [M

bps]

k-shortest MDD MDD-LF k-shortest MDD MDD-LF0

0.2

0.4

0.6

0.8

1

Max

imum

link

load

Compared to k-shortest, MDD: 1.7 times higher load MDD-LF: 2.3 times higher load

MDD and MDD-LF tend to select longer hop paths than k-shortest

Page 18: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Evaluation of TCP Performance over Fixed Evaluation of TCP Performance over Fixed Robust Routing: ResultRobust Routing: Result Larger delay difference

more candidates for overtaking packet Higher ratio of shorter path

higher probability of three out-of-order packet arrivals

SLLL: 0.012 SSLL: 0.063 SSSL: 0.11

CQR 2011 18

d (delay difference between path L and path S)

0

20

40

60

80

100

0 0.4 0.8 1.2 1.6 2 2.4 2.8

Tota

l th

rou

gh

pu

t [M

bp

s]

d [ms]0.4 0.8 1.2 1.6 2.0 2.4 2.80

0

20

40

60

80

100

Tot

al t

hro

ughp

ut [

Mbp

s]SLLL

SSLL

SSSL

Lower TCP throughput

Average packet transmission interval

Page 19: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Proposal of Fixed Robust Routing Taking Account of TCP Proposal of Fixed Robust Routing Taking Account of TCP Performance (2/2): AlgorithmPerformance (2/2): Algorithm

19

Step. 1 Selection of candidate paths of every source-destination pair Step. 1.1 We select K shortest hop paths Step. 1.2 From the K paths, we select M paths with the minimum delay difference between the shortest and the longest delay pathsStep. 2. We solve the formulated problem and obtain maximum link load (t) and fraction (xp) of traffic routed on every path. In MDD-LF, we bound fraction of traffic routed on the shortest delay path by α

Path 1, 15ms

Path 2, 8ms

Path 3, 3ms

Path 4, 14ms

Path 5, 10ms

MDD (Minimum Delay Difference) MDD-LF (MDD with Limited Fraction)and

Node i Node j

Page 20: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Simulation ModelSimulation Model

One TCP connection for every node-pair (Ri , Rj) Each source host’s data transmission rate: 10 [Mbps] Parameter settings in MDD and MDD-LF

K = 5 M = 2 α = 0.25

Comparison: k-shortest A straightforward fixed robust routing that selects M (= 2) shortest hop paths as candidate paths for every node-pair

CQR 2011 20

R4[2%]

4.7ms

2.8ms

7.0ms

3.5ms 2.8ms 3.5ms

3.5ms

3.5ms

3.5ms3.5ms

8.4ms

8.4ms

4.9ms

5.6ms

11.2ms

2.8ms

9.1ms

0.7ms

1.4ms

1.4ms

1.4msLink bandwidth: 1 [Gbps]

Page 21: Performance Evaluation  of TCP over Multiple Paths  in Fixed Robust Routing

Conclusions and Future WorkConclusions and Future Work

Conclusions Investigation of TCP throughput over fixed robust routing

Larger delay difference Higher ratio of shorter path

Proposal of fixed robust routing algorithms that try to improve TCP throughput

MDD: 22% higher throughput but 1.7times higher load MDD-LF: 27% higher throughput but 2.3 times higher load

Future work Performance evaluation of our proposed algorithms in detail Modification of our proposed algorithms

Selection of link-disjoint paths as candidate paths

CQR 2011 21

Lower TCP throughput