Minimum-Delay Load-Balancing Through Non-Parametric Regression F. Larroca and J.-L. Rougier IFIP/TC6 Networking 2009 Aachen, Germany, 11-15 May 2009.

Slides:



Advertisements
Similar presentations
Ch. 12 Routing in Switched Networks
Advertisements

Routing and Congestion Problems in General Networks Presented by Jun Zou CAS 744.
Price Of Anarchy: Routing
How Bad is Selfish Routing? By Tim Roughgarden Eva Tardos Presented by Alex Kogan.
Dynamic Traffic Engineering Techniques for the Internet Federico Larroca Supervisor: Jean-Louis Rougier December 18th 2009.
1 EL736 Communications Networks II: Design and Algorithms Class8: Networks with Shortest-Path Routing Yong Liu 10/31/2007.
Achieving Network Optima Using Stackelberg Routing Strategies Yannis A. Korilis, Member, IEEE Aurel A. Lazar, Fellow, IEEE & Ariel Orda, Member IEEE IEEE/ACM.
Pricing Granularity for Congestion-Sensitive Pricing Murat Yüksel and Shivkumar Kalyanaraman Rensselaer Polytechnic Institute, Troy, NY {yuksem, shivkuma}
Competitive Routing in Multi-User Communication Networks Presentation By: Yuval Lifshitz In Seminar: Computational Issues in Game Theory (2002/3) By: Prof.
Charge-Sensitive TCP and Rate Control Richard J. La Department of EECS UC Berkeley November 22, 1999.
Beyond selfish routing: Network Formation Games. Network Formation Games NFGs model the various ways in which selfish agents might create/use networks.
Minimaxity & Admissibility Presenting: Slava Chernoi Lehman and Casella, chapter 5 sections 1-2,7.
1 Chapter 8: Linearization Methods for Constrained Problems Book Review Presented by Kartik Pandit July 23, 2010 ENGINEERING OPTIMIZATION Methods and Applications.
A Scalable Network Resource Allocation Mechanism With Bounded Efficiency Loss IEEE Journal on Selected Areas in Communications, 2006 Johari, R., Tsitsiklis,
EE 685 presentation Optimization Flow Control, I: Basic Algorithm and Convergence By Steven Low and David Lapsley Asynchronous Distributed Algorithm Proof.
Rethinking Internet Traffic Management: From Multiple Decompositions to a Practical Protocol Jiayue He Princeton University Joint work with Martin Suchara,
1 A Novel Mechanism for Flooding Based Route Discovery in Ad hoc Networks Jian Li and Prasant Mohapatra Networks Lab, UC Davis.
How bad is selfish routing? מאת ליאור עזרא עפ " י מאמר של חווה טרדוס וטים גן - מחוספס.
Analysis of the Increase and Decrease Algorithms for Congestion Avoidance in Computer Networks Dah-Ming Chiu and Raj Jain Presented by Aaron Ballew (slides.
Multipath Protocol for Delay-Sensitive Traffic Jennifer Rexford Princeton University Joint work with Umar Javed, Martin Suchara, and Jiayue He
On Self Adaptive Routing in Dynamic Environments -- A probabilistic routing scheme Haiyong Xie, Lili Qiu, Yang Richard Yang and Yin Yale, MR and.
A Criticism of: “Moving beyond end-to-end path information to optimize CDN performance” Gautam Bhawsar Alok Rakkhit.
1 A Cooperative Game Framework for QoS Guided Job Allocation Schemes in Grids Riky Subrata, Member, IEEE, Albert Y. Zomaya, Fellow, IEEE, and Bjorn Landfeldt,
Routing Games for Traffic Engineering F. Larroca and J.L. Rougier IEEE International Conference on Communications (ICC 2009) Dresden, Germany, June
MATE: MPLS Adaptive Traffic Engineering Anwar Elwalid, et. al. IEEE INFOCOM 2001.
1 Topology Control of Multihop Wireless Networks Using Transmit Power Adjustment Infocom /12/20.
Flow Models and Optimal Routing. How can we evaluate the performance of a routing algorithm –quantify how well they do –use arrival rates at nodes and.
Robust Regression for Minimum-Delay Load-Balancing F. Larroca and J.-L. Rougier 21st International Teletraffic Congress (ITC 21) Paris, France, September.
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jennifer Rexford Princeton University With Jiayue He, Rui Zhang-Shen, Ying Li,
Particle Filtering in Network Tomography
L13. Shortest path routing D. Moltchanov, TUT, Spring 2008 D. Moltchanov, TUT, Spring 2014.
1 Mohammed M. Olama Seddik M. Djouadi ECE Department/University of Tennessee Ioannis G. PapageorgiouCharalambos D. Charalambous Ioannis G. Papageorgiou.
NOBEL WP Szept Stockholm Game Theory in Inter-domain Routing LÓJA Krisztina - SZIGETI János - CINKLER Tibor BME TMIT Budapest,
A Fair and Dynamic Load Balancing Mechanism F. Larroca and J.L. Rougier International Workshop on Traffic Management and Traffic Engineering for the Future.
CSC321: Neural Networks Lecture 12: Clustering Geoffrey Hinton.
NETE4631:Capacity Planning (2)- Lecture 10 Suronapee Phoomvuthisarn, Ph.D. /
Dynamics of Traffic Flows in Combined Day-to-day and With- in Day Context Chandra Balijepalli ITS, Leeds September 2004.
Interaction of Overlay Networks: Properties and Implications Joe W.J. Jiang Dah-Ming Chiu John C.S. Lui The Chinese University of Hong Kong.
Peter Key Cambridge UK joint work with Richard Gibbens, Statistical Laboratory, Cambridge Uni. UK The.
Michael Schapira Yale and UC Berkeley Joint work with P. Brighten Godfrey, Aviv Zohar and Scott Shenker.
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jiayue He, Rui Zhang-Shen, Ying Li, Cheng-Yen Lee, Jennifer Rexford, and Mung.
EE 685 presentation Optimization Flow Control, I: Basic Algorithm and Convergence By Steven Low and David Lapsley.
Network Congestion Games
1 - CS7701 – Fall 2004 Review of: Detecting Network Intrusions via Sampling: A Game Theoretic Approach Paper by: – Murali Kodialam (Bell Labs) – T.V. Lakshman.
Beyond selfish routing: Network Games. Network Games NGs model the various ways in which selfish agents strategically interact in using a network They.
June 4, 2003EE384Y1 Demand Based Rate Allocation Arpita Ghosh and James Mammen {arpitag, EE 384Y Project 4 th June, 2003.
Designing Games for Distributed Optimization Na Li and Jason R. Marden IEEE Journal of Selected Topics in Signal Processing, Vol. 7, No. 2, pp ,
1  The Problem: Consider a two class task with ω 1, ω 2   LINEAR CLASSIFIERS.
Algorithms for Resource Allocation in HetNet Jianwei Liu Clemson University.
QoS Routing and Scheduling in TDMA based Wireless Mesh Backhaul Networks Chi-Yao Hong, Ai-Chun Pang,and Jean-Lien C. Wu IEEE Wireless Communications and.
1 Slides by Yong Liu 1, Deep Medhi 2, and Michał Pióro 3 1 Polytechnic University, New York, USA 2 University of Missouri-Kansas City, USA 3 Warsaw University.
Managing Congestion and Emissions in Road Networks with Tolls and Rebates Hai Yang Chair Professor Department of Civil and Environmental Engineering The.
Tunable QoS-Aware Network Survivability Presenter : Yen Fen Kao Advisor : Yeong Sung Lin 2013 Proceedings IEEE INFOCOM.
2/14/2016  A. Orda, A. Segall, 1 Queueing Networks M nodes external arrival rate (Poisson) service rate in each node (exponential) upon service completion.
MATE: MPLS Adaptive Traffic Engineering Anwar Elwalid Cheng Jin Steven Low Indra Widjaja Bell Labs Michigan altech Fujitsu 2006.
Generated Trips and their Implications for Transport Modelling using EMME/2 Marwan AL-Azzawi Senior Transport Planner PDC Consultants, UK Also at Napier.
Linear Programming Chapter 1 Introduction.
Travel Demand Forecasting: Traffic Assignment CE331 Transportation Engineering.
1 Traffic Engineering By Kavitha Ganapa. 2 Introduction Traffic engineering is concerned with the issue of performance evaluation and optimization of.
D. AriflerCMPE 548 Fall CMPE 548 Routing and Congestion Control.
Load Balancing of Multipath Source Routing in Ad Hoc Networks Lianfang Zhang, Zenghua Zhao, Yantai Shu, and Lei Wang Department of Computer Science Tianjin.
Virtual-Topology Adaptation for WDM Mesh Networks Under Dynamic Traffic.
Impact of Interference on Multi-hop Wireless Network Performance
Deep Feedforward Networks
One-layer neural networks Approximation problems
URBAN TRANSPORTATION NERWORKS
Rutgers Intelligent Transportation Systems (RITS) Laboratory
James B. Orlin Presented by Tal Kaminker
Advisor: Frank Yeong-Sung Lin, Ph.D. Presented by Yu-Jen Hsieh 謝友仁
Resource Pooling A system exhibits complete resource pooling if it behaves as if there was a single pooled resource. I propose ‘extent of resource pooling’
Presentation transcript:

Minimum-Delay Load-Balancing Through Non-Parametric Regression F. Larroca and J.-L. Rougier IFIP/TC6 Networking 2009 Aachen, Germany, May 2009

page 1 Introduction Current traffic is highly dynamic and unpredictable How may we define a routing scheme that performs well under these demanding conditions? Possible Answer: Dynamic Load-Balancing We connect each Origin-Destination (OD) pair with several pre-established paths Traffic is distributed in order to optimize a certain function Function f l (  l ) is typically a convex increasing function that diverges as  l → c l ; e.g. mean queuing delay Why queuing delay? Simplicity and versatility IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 2 Introduction A simple model (M/M/1) is always assumed What happens when we are interested in actually minimizing the total delay? Simple models are inadequate We propose: Make the minimum assumptions on f l (  l ) (e.g. monotone increasing) Learn it from measurements instead Optimize with this learnt function IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 3 Agenda Introduction Attaining the optimum Delay function approximation Simulations Conclusions IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 4 Problem Definition Queuing delay on link l is given by D l (  l ) Our congestion measure: weighted mean end-to-end queuing delay The problem: Since f l (  l ):=  l D l (  l ) is proportional to the queue size, we will use this value instead IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 5 Congestion Routing Game Path P has an associated cost  P : where  l (  l ) is continuous, positive and non-decreasing Each OD pair greedily adjusts its traffic distribution to minimize its total cost Equilibrium: no OD pair may decrease its total cost by unilaterally changing its traffic distribution It coincides with the minimum of: IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 6 Congestion Routing Game What happens if we use ? The equilibrium coincides with the minimum of: To solve our problem, we may play a Congestion Routing Game with To converge to the Equilibrium we will use REPLEX Important:  l (  l ) should be continuous, positive and non-decreasing IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 7 Agenda Introduction Attaining the optimum Delay function approximation Simulations Conclusions IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 8 Cost Function Approximation What should be used as f l (  l )? 1.That represents reality as much as possible 2.Whose derivative (  l (  l )) is: a.continuous b.positive => f l (  l ) non-decreasing c.non-decreasing => f l (  l ) convex To address 1 we estimate f l (  l ) from measurements Convex Nonparametric Least-Squares (CNLS) is used to enforce 2.b and 2.c : Given a set of measurements {(  i,Y i )} i=1,..,N find f N F where F is the set of continuous, non-decreasing and convex functions IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 9 Cost Function Approximation The size of F complicates the problem Consider instead G (subset of F) a family of piecewise- linear convex non-decreasing functions The same optimum is obtained if we change F by G We may now rewrite the problem as a standard QP one IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 10 Cost Function Approximation This regression function presents a problem: its derivative is not continuous (cf. 2.b) A soft approximation of a piecewise linear function: Our final approximation of the link-cost function: IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 11 An Example IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 12 Agenda Introduction Attaining the optimum Delay function approximation Simulations Conclusions IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 13 NS-2 simulations The considered network: IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 14 NS-2 simulations Alternative (“wrong”) training set: IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 15 Agenda Introduction Attaining the optimum Delay function approximation Simulations Conclusions IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 16 Conclusions and Future Work We have presented a framework to converge to the actual minimum total mean delay demand vector Two shortcomings of our framework:  l (  l ) is constant outside the support of the observations Links with little or no queue size have a negligible cost Possible Solution: Add a “patch” function that is negligible with respect to  l (  l ) except at high loads How does  l (  l ) behaves over time? Does it change? How often? How does our framework performs when compared with other mechanisms or simpler models? Faster and/or more robust alternative regression methods? Is REPLEX the best choice? IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier

page 17IFIP/TC6 Networking 2009 F. Larroca and J.-L. Rougier Thank you! Questions?