Network Layer Support for Gigabit TCP Flows in Wireless Mesh - - PowerPoint PPT Presentation

network layer support for gigabit tcp flows in wireless
SMART_READER_LITE
LIVE PREVIEW

Network Layer Support for Gigabit TCP Flows in Wireless Mesh - - PowerPoint PPT Presentation

Network Layer Support for Gigabit TCP Flows in Wireless Mesh Networks This work is collaborated with Prof. Ramanathan, Univ. of Wisconsin Madison Chin-ya Huang Feb. 6, 2015 1 BBBBBBBBBBBBBBBB Outline Motivation Challenges


slide-1
SLIDE 1

BBBBBBBBBBBBBBBB Network Layer Support for Gigabit TCP Flows in Wireless Mesh Networks

This work is collaborated with

  • Prof. Ramanathan, Univ. of Wisconsin Madison

Chin-ya Huang

  • Feb. 6, 2015

1

slide-2
SLIDE 2

2

Outline

  • Motivation
  • Challenges

– Dynamic change of network environment – Related works

  • Spare Bandwidth Rate Adaptive Network Coding (SRNC) Scheme

– Digraph diversity – Inter-flow network coding – Spare bandwidth exploitation – Buffer management

  • Evaluation
  • Conclusions and Discussions

– Summary of SRNC – Extended research of SRNC

2

slide-3
SLIDE 3

3

Motivation

  • Growing need of real-time streaming applications :

– HDTV – Video conferencing – Movie on demands

  • Characteristics of these applications [Goel et al, TOMCCAP 2008] :

– Long-lived – High data rate for each application (>1Gbps in the future)

  • Seamless deliver these applications require

– High physical layer data rate (>1Gbps) – New techniques and protocols at higher layers

3

slide-4
SLIDE 4

4

Problems of Mesh Networks

  • Mesh networks are able to provide high data link rate

– Gigamesh network provides gigabit physical layer data rate in mesh networks – When data rate exceeds Gbps, packet loss, handoff delays, re-routing,

  • etc. would have more severe impact on network throughput

4

slide-5
SLIDE 5

5

Impact of different link bandwidth

  • TCP throughput drops while the link bandwidth fluctuates.

– When the link bandwidth exceeds 1Gpb, TCP throughput < 50% of the maxflow.

5

0.75 0.59 0.31 0.21 0.21 0.74 0.63 0.46 0.42 0.27 0.99 0.99 0.99 0.99 0.99 0.2 0.4 0.6 0.8 1 1.2 10M 100M 500M 800M 1G Normalized TCP Throughput Bottleneck link bandwidth (bps) NewReno CTCP CBR

H I 10 TCP flows 10 TCP flows Cross traffic G C B A

slide-6
SLIDE 6

6

Spare Bandwidth Rate Adaptive Network Coding (SRNC)

[Huang et al, TMC 2014]

Goal: To develop strategies for improving end-to-end throughput – TCP based

Develop and integrate the following four ideas: – Digraph diversity (direct acyclic graph diversity) – Inter-flow network coding – Spare bandwidth exploitation – Buffer management

6

slide-7
SLIDE 7

7

  • Forward packets through

multiple routes [Wang et al,

CoRoNet 2009]

Pro: + More tolerant congestion + Load balance Con:

  • Routing is more

expensive

  • Fully exploit rerouting

Related Work: multipath routing A B C E F G H

7

slide-8
SLIDE 8

8

Related Work: forward error correction (FEC)

  • Exploit available

bandwidth by sending redundant packets to recover packet loss [Medard

et al, INFOCOM 2009]

+ Loss tolerance

  • Require feedback of

network condition

  • Require sophisticated

schemes to adaptively exploit available bandwidth

  • Increase packet

losses due to buffer

  • verflow

1 1 3 3 1 2 1 2 1 A B C E F H G 2 2 3 2 2 1 2

8

slide-9
SLIDE 9

9

Motivation of SRNC

  • SRNC: Adapt the idea of network coding into unicast transmission

along with previous ideas

  • Network coding: linear combination of packets in the queue
  • Network coding is first proposed for multicasting aimed to increase

datarate in 2000 [Ahlswede et al, TIT 2000].

  • Packets are encoded from the same flow [Medard et al, INFOCOM 2009].
  • Specifically, we network encode packets from multiple flows.

We will show that: + Adapts to the changes in available bandwidth in a distributed fashion + Needs much less re-routing

9

slide-10
SLIDE 10

10

Outline

  • Motivation
  • Challenges

– Dynamic change of network environment – Related works

  • Spare Bandwidth Rate Adaptive Network Coding (SRNC) Scheme

– Digraph diversity – Inter-flow network coding – Spare bandwidth exploitation – Buffer management

  • Evaluation
  • Conclusions and Discussions

– Summary of SRNC – Extended research of SRNC

10

slide-11
SLIDE 11

11

  • Gateway node (A) finds the direct graph in the network for multipath routing

based on the destination (D1, D2).

SRNC: digraph diversity

2 2 1 1 3 3 2 2 1 D2 S1 S2 D1

A B C D E F G

11

slide-12
SLIDE 12

12

SRNC: inter-flow network coding

2 2 2 2 3 3 2 2 1 D2 S1 S2 D1

A B C D E F G

  • Packets are encoded from multiple flows.
  • All mesh nodes network encode packets before forwarding them.
  • Loss of one packet may result in loss of all coded packets.

12

slide-13
SLIDE 13

13

SRNC: exploit available bandwidth

  • Assume all traffic from A to G without cross traffic
  • Each mesh node conditionally sends as many as possible to next hop

– Send additional packets to outgoing links in a distributed fashion 2 2 2 2 3 3 2 2 1 D2 S1 S2 D1

A B C D E F G

13

slide-14
SLIDE 14

14

SRNC: buffer management

  • Gateway marks appropriate number of packets as high priority
  • Each mesh node adaptively manages packets in the queue.

– High priority packets are sent first – Forward high priority packets no more than received – Drop low priority from buffer if necessary H H H L L

a) More bandwidth on outgoing links

H H H L L

b) Buffer overflow occurs

14

slide-15
SLIDE 15

15

SRNC: example

Packets are received successfully!

  • Although packet loss occurs, packets are still decodable because

SRNC fully and distribultedly utilize the available bandwidth for transmission aimed to recover the packet loss during transmission.

2 2 2 2 3 3 2 2 1 D2 S1 S2 D1

A B C D E F G

15

slide-16
SLIDE 16

16

SRNC

  • Packets are routed in a routing digraph.
  • Packets across different flows are linearly combined before being forwarded to

each mesh node’s outgoing links.

  • Each mesh node forwards network encoded packets based on the available

bandwidth. – Number of forwarded packets is not necessary to be the same as received. – Packets are discarded conditionally when necessary.

  • Access node reconstructs packets after decoding.

2 2 2 2 3 3 2 2 1 D2 S1 S2 D1 G C F H B E A

16

slide-17
SLIDE 17

17

Outline

  • Motivation
  • Challenges

– Dynamic change of network environment – Related works

  • Spare Bandwidth Rate Adaptive Network Coding (SRNC) Scheme

– Inter-flow network coding – Diagraph routing – Forward error correction – Buffer management

  • Evaluation
  • Conclusions and Discussions

– Summary of SRNC – Extended research of SRNC

17

slide-18
SLIDE 18

18

Simulation Configurations

x (Mbps) : Average link bandwidth in the mesh. y (Mbps) : Link bandwidth to the users. Simulation tool: Network Simlator-2 (NS-2) Simulation topology:

18

H A B C D G I J x x x x x x x x x x y y y y Internet : Mesh node : TCP sender : Wireless host

slide-19
SLIDE 19

19

Impact of different link bandwidth

  • Maximum possible TCP Throughput = min {y, }
  • x = 10 Gbps

19

Large Spare Bandwidth Moderate Spare Bandwidth Tight Spare Bandwidth 100 200 300 400 500 600 700 800 900 1000 500M 800M 1G 1.2G 1.5G Average TCP Throughput (Mbps) MR PRO [JK2009] SRNC Link bandwidth between mobile hosts and

slide-20
SLIDE 20

20

0.5 1 1.5 2 TCP Cross-traffic Before After Ratio of SRNC over MR Throughput

Fairness between heterogeneous flow sets

  • TCP Cross-traffic does not implement SRNC.
  • The performance of cross-traffic is not effected by SRNC.
  • The performance of TCP in SRNC is improved after bandwidth fluctuation.

– SRNC utilizes spare bandwidth to enhance the throughput.

20

slide-21
SLIDE 21

21

Effectiveness in Different Topologies

: TCP flows : TCP Cross-traffic flows

21

(a) Topo 1 (b) Topo 2 (c) Topo 3 (d) Topo 4 (e) Topo 5 (f) Topo 6 (g) Topo 7

slide-22
SLIDE 22

22

Effectiveness in Different Topologies

(TCP,Before) (TCP,After) (Cross- traffic,Before) (Cross- traffic,After) Topo 1 0.93 1.98 1 1.05 Topo 2 0.95 5.25 1 1.04 Topo 3 0.95 2.15 1 1 Topo 4 0.9 3.61 1 1.09 Topo 5 0.93 9.11 1 1.36 Topo 6 0.93 4.23 1 1.48 Topo 7 0.93 0.72 1 1.43

22

The achieved throughput for TCP and TCP cross-traffic before and after bandwidth fluctuation

slide-23
SLIDE 23

23

Impact of wireless loss

  • Consider wireless loss occurs on each link with independent probability p.
  • Approximately, the wireless loss of the mesh is 4p.

23

100 200 300 400 500 600 700 2 4 6 8 10 TCP Throughput (Mbps) Overall Wireless Packet Loss Probability (%) PRO [JK2009] SRNC

slide-24
SLIDE 24

24

Outline

  • Motivation
  • Challenges

– Dynamic change of network environment – Related works

  • Spare Bandwidth Rate Adaptive Network Coding (SRNC) Scheme

– Inter-flow network coding – Diagraph routing – Forward error correction – Buffer management

  • Evaluation
  • Conclusions and Discussions

– Summary of SRNC – Extended research of SRNC

24

slide-25
SLIDE 25

25

Summary

  • Link bandwidth fluctuates due to the change of network

environments. – TCP throughput drops significantly when BDP increases. – Network needs to react to the change to sustain the performance.

  • SRNC increases TCP throughput by effectively utilizing available

bandwidth in the network. – Four components are involved.

  • SRNC shares bandwidth fairly with cross traffic no matter SRNC

is applied or not.

  • SRNC is fully distributed and can be integrated in existing

protocol stack and deploying in real network testbed.

25

slide-26
SLIDE 26

26

Experimenting with SRNC

  • To execute the network coding in a link rate

– Implement on FPGA

  • Global Environment for Network Innovations (GENI) project :

– Implementing SRNC in Openflow enabled router. – Evaluating SRNC in network testbed.

26

Internet Internet

slide-27
SLIDE 27

27

Extended research of SRNC

  • One characteristic of gigabit TCP flows in the mesh is large bandwidth delay

product (BDP).

  • Underwater mesh networks also have large BDP.

– The BDP between gigabits mesh and underwater mesh is similar. – Apply the idea of SRNC to improve the end-to-end throughput in the underwater mesh networks [Huang et al, JSAC, 2011].

27

slide-28
SLIDE 28

28

Other Applications of SRNC

  • Cognitive radio enabled networks
  • Data centers
  • Priority based traffic management for real-time flows
  • Non-TCP type flows
  • Anything else ?

28

slide-29
SLIDE 29

29

29

slide-30
SLIDE 30

30

Thank you for your attention ! Please feel free to contact me by e-mail : yaya.wisc@gmail.com if you have any thoughts regarding this work.

30