distributed traffic management framework

Distributed Traffic Management Framework

Saurabh NambiarSuryaPrabha C P

Smruthi KShijil

Department of Computer Science andEngineering

Govt. College of Engineering, Kannur

Under the Guidance ofProf. Najeeb K

1 of 35

Problem Statement

Distributed traffic management framework in which routers aredeployed with intelligent data rate controllers to tackle traffic mass inhigh speed networks.

2 of 35

Project Outline

• Outline

• Introduction

• Background Information

• Literature Survey

• Requirement and Specification

• Proposed Work

• Design

• Implementation

• Snapshots

• Conclusion

• References

3 of 35

IntroductionMotivation

• To tackle congestion in high speed traffic.

• To implement fuzzy logic in network.

4 of 35

IntroductionPurpose and Goal

1. Using fuzzy logic theory to design an explicit rate-based trafficmanagement scheme.

2. The application of fuzzy logic controller using less performanceparameters while providing better performances than the existing.

3. The design of a Fuzzy Smoother mechanism that generaterelatively smooth flow throughput.

5 of 35

Background Information

• Congestion control

• Fuzzy logic control

• Quality of service

• Max-Min fairness

• Robustness

• Traffic management

6 of 35

Literature Survey

• TCP feature - source adjusts its window size based on packet losssignal.

• TCP encounters various performance problems ,when the InternetBDP continues to increase.

1. Utilization2. Fairness3. Stability

7 of 35

Literature Survey

• Explicit Congestion Controls Protocols:

1. XCP2. RCP3. JetMax4. MaxNet

• Inaccurate estimation resulting in performance degradation.

• Queue size not stable due to oscillations- affects the stability oftheir sending rates.

8 of 35

HARDWARE REQUIREMENTS

• Processor - Pentium IV

• Speed - 1.1 Ghz

• RAM - 256 MB(min)

• Hard Disk - 20 GB

• Key Board - Standard Windows Keyboard

• Mouse - Two or Three Button Mouse

• Monitor - SVGA

9 of 35

SOFTWARE REQUIREMENTS

• Operating System : Windows XP

• Programming Language : JAVA.

• Java Version : JDK 1.6 & above.

10 of 35

Proposed work

• Using fuzzy logic theory to design an explicit rate-based trafficmanagement scheme (called the IntelRate controller) for thehigh-speed IP networks.

• The application of such a fuzzy logic controller using lessperformance parameters while providing better performances thanthe existing explicit traffic control protocols.

• The design of a Fuzzy Smoother mechanism that can generaterelatively smooth flow throughput.

11 of 35

DesignNetwork Model

12 of 35

Network Model

• Congestion occurs when IQSize exceeds buffer capacity.

• Distributed traffic controller implemented inside each router.

• Req rate : Stores the sending rate of source

• Routers role:◦ Calculates source sending rate according to IQSize.◦ Compares it with Req rate.◦ Modifies this field to the lowest of two values.

• Modified value is sent to source using Acknowledgementpacket(ACK).

13 of 35

Network Model

• For a particular source-destination pair,RTPD, τpi = τfi1 + τfi2 + τbi

RTT, τi = Propagation delay + queueing delay + processing delay

• Let q(t) be the router queue size(IQSize)if q(t) > 0, q(t) = y(t) + v(t) − c(t)if q(t) = 0, q(t) = [y(t) + v(t) − c(t)]+ where [x ]+ = max(0, x)

14 of 35

IntelRate Controller Design

• TISO(Two Input Single Output)

• Queue Deviation(error), e(t) = q0 - q(t)

• To remove steady state error, g(e(t)) =∫e(t).dt

• Aggeregate output, y(t) = Σui (t − τi )

15 of 35

IntelRate Controller DesignFuzzy Logic Control

• FLC is non linear mapping of inputs to outputs.

• Four parts:

1. Rule base building2. Fuzzification3. Inference4. Defuzzification

16 of 35

Fuzzy logic controlRule Base building

• Rule base is set of linguistic values used to map inputs to outputsusing ”If...Then” format.

Figure : Rule base of IntelRate Controller

17 of 35

Fuzzy Logic ControlFuzzification

• Transforming crisp values into grades of membership of fuzzy set.

• Membership functions(MF) are used for this transformation.

• Fuzzifier: triangular or trapezoidal.

• For any two inputs P1andP2, certainty of a rule is given byZadeh’s AND Logic:µP1

m(p1)⋂µP2

m(p2) = min(µP1m(p1), µP2

m(p2)) : p[i]εPi

18 of 35

Fuzzy Logic ControlInference with Fuzzy Smoother

• Fuzzy smoother:

1. realize a smaller TBO.2. reduce queueing delay upon heavy traffic.

• Upper and lower limits are set for both inputs.

1. −mq0 ≤ g(e(t)) ≤ mq0

2. q0 − B ≤ e(t) ≤ q0

19 of 35

Fuzzy Logic ControlDe-Fuzzification

• Membership degree of fuzzy set is transformed to real valued result.

• IntelRate controller uses COG(Centre of Gravity) method.u(t) = (ΣcjSj/(ΣSj )), j = 1, 2..k

where k = No. of rulescj is bottom centroid of triangular MFSj is the area of triangle

20 of 35

ImplementationSimulation Model

Figure : Simulated Network

21 of 35

MODULES

• Sender

• Receiver

• Router Queuing Scheme

22 of 35

SENDER

• Main module

• Divides the file into packets.

• Requests sending rate it desires by depositing a value into adedicated field Req rate inside the packet header.

• Message log to store all requests and references made.

23 of 35

RECEIVER

• The receiver then sends this value back to the source via an ACKpacket.

• Source modifies the sending rate accordingly.

• Feedback by receiver helps in congestion control.

24 of 35

ROUTER QUEUING SCHEME

• In this module,router computes transmission rate based on IQ Size

• Compare it with the rate already recorded in Req rate field.

• Chooses lowest value among them.

• Network analysis is done to enhance performance parameters.

25 of 35

SnapshotsSource Node

26 of 35

SnapshotsRouter

27 of 35

SnapshotsRouter Analysis

28 of 35


29 of 35


30 of 35

SnapshotsReceiver Node

31 of 35

Conclusion

• IntelRate controller manages the Internet congestion in order toassure the quality of service for different service applications.

• The controller is designed to improve upon the disadvantages ofearlier used congestion protocols.

• Use of Fuzzy logic provides the intelligence equivalent to humansfor decision making.

32 of 35

Future Work

• Enhancements are possible for a faster design , that is, fasterqueue size calculation using sophisticated techniques.

• In low speed networks, this framework can be enhanced forefficient utilization of bandwidth.

• Better congestion protocols can be implemented using thisframework so that it is compatible with TCP(widely used currentand future transport layer protocol).

33 of 35

References

• IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, VOL. 10, NO. 2, JUNE 2013–UsingFuzzy Logic Control to Provide Intelligent Traffic Management Service for High-Speed Networks

• 11th International Conference on Telecommunications - ConTEL 2011 ISBN: 978-953-184-152-8, June 15-17,2011, Graz, Austria–Fuzzy CAC based Traffic Management

• International Journal of Advances in Engineering & Technology, Nov 2011–DESIGN AND SIMULATION OF ANINTELLIGENT TRAFFIC CONTROL SYSTEM

• J. Liu and O. Yang, Stability analysis and evaluation of the IntelRate controller for high-speed heterogeneousnetworks, in Proc. 2011 IEEE ICC, pp. 1-5

34 of 35

Thank You ! ! !

35 of 35