Building a Strong Foundation for a Future Internet Jennifer Rexford ’91 Computer Science Department (and Electrical Engineering and the Center for IT Policy)

Slides:



Advertisements
Similar presentations
Internet Economics: the use of Shapley value for ISP settlement Richard T.B. Ma Columbia University Dah-ming Chiu, John C.S. Lui The Chinese University.
Advertisements

Network Layer: Internet-Wide Routing & BGP Dina Katabi & Sam Madden.
Fundamentals of Computer Networks ECE 478/578 Lecture #18: Policy-Based Routing Instructor: Loukas Lazos Dept of Electrical and Computer Engineering University.
Consensus Routing: The Internet as a Distributed System John P. John, Ethan Katz-Bassett, Arvind Krishnamurthy, and Thomas Anderson Presented.
Lecture 6 Overlay Networks CPE 401/601 Computer Network Systems slides are modified from Jennifer Rexford.
Towards a Logic for Wide-Area Internet Routing Nick Feamster and Hari Balakrishnan M.I.T. Computer Science and Artificial Intelligence Laboratory Kunal.
Part II: Inter-domain Routing Policies. March 8, What is routing policy? ISP1 ISP4ISP3 Cust1Cust2 ISP2 traffic Connectivity DOES NOT imply reachability!
Can Economic Incentives Make the ‘Net Work? Jennifer Rexford Princeton University
Game Theoretic and Economic Perspectives on Interdomain Routing Michael Schapira Yale University and UC Berkeley.
CPSC Topics in Multimedia Networking A Mechanism for Equitable Bandwidth Allocation under QoS and Budget Constraints D. Sivakumar IBM Almaden Research.
TIE Breaking: Tunable Interdomain Egress Selection Renata Teixeira Laboratoire d’Informatique de Paris 6 Université Pierre et Marie Curie with Tim Griffin.
Can Congestion Control and Traffic Engineering be at Odds? Jiayue He, Mung Chiang, Jennifer Rexford Princeton University November 30 th, 2006.
1 In VINI Veritas: Realistic and Controlled Network Experimentation Jennifer Rexford with Andy Bavier, Nick Feamster, Mark Huang, and Larry Peterson
Traffic Engineering With Traditional IP Routing Protocols
1 Tutorial 5 Safe “Peering Backup” Routing With BGP Based on:
Mini Introduction to BGP Michalis Faloutsos. What Is BGP?  Border Gateway Protocol BGP-4  The de-facto interdomain routing protocol  BGP enables policy.
Traffic Engineering Jennifer Rexford Advanced Computer Networks Tuesdays/Thursdays 1:30pm-2:50pm.
Internet Networking Spring 2004 Tutorial 5 Safe “Peering Backup” Routing With BGP.
Rethinking Traffic Management: Using Optimization Decomposition to Derive New Architectures Jennifer Rexford Princeton University Jiayue He, Ma’ayan Bresler,
1 GENI: Global Environment for Network Innovations Jennifer Rexford On behalf of Allison Mankin (NSF)
Rethinking Internet Traffic Management Using Optimization Theory Jennifer Rexford Princeton University Joint work with Jiayue He, Martin Suchara, Ma’ayan.
New Routing Architectures Jennifer Rexford Advanced Computer Networks Tuesdays/Thursdays 1:30pm-2:50pm.
Stable Internet Routing Without Global Coordination Jennifer Rexford Princeton University Joint work with Lixin Gao (UMass-Amherst)
Slide -1- February, 2006 Interdomain Routing Gordon Wilfong Distinguished Member of Technical Staff Algorithms Research Department Mathematical and Algorithmic.
Future Research Directions Jennifer Rexford Advanced Computer Networks Tuesdays/Thursdays 1:30pm-2:50pm.
Interdomain Routing Establish routes between autonomous systems (ASes). Currently done with the Border Gateway Protocol (BGP). AT&T Qwest Comcast Verizon.
Inherently Safe Backup Routing with BGP Lixin Gao (U. Mass Amherst) Timothy Griffin (AT&T Research) Jennifer Rexford (AT&T Research)
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 Exterior Gateway Protocols: EGP, BGP-4, CIDR Shivkumar Kalyanaraman Rensselaer Polytechnic Institute.
Economic Incentives in Internet Routing Jennifer Rexford Princeton University
Rethinking Internet Traffic Management: From Multiple Decompositions to a Practical Protocol Jiayue He Princeton University Joint work with Martin Suchara,
Stable Internet Routing Without Global Coordination Jennifer Rexford AT&T Labs--Research
1 Interdomain Routing Policy Reading: Sections plus optional reading COS 461: Computer Networks Spring 2008 (MW 1:30-2:50 in COS 105) Jennifer Rexford.
Building a Strong Foundation for a Future Internet Jennifer Rexford Princeton University
Backbone Networks Jennifer Rexford COS 461: Computer Networks Lectures: MW 10-10:50am in Architecture N101
1 Algorithmic Foundations of the Internet Alejandro Lopez-Ortiz (Waterloo) Jennifer Rexford (Princeton) Rebecca Wright (Stevens)
Looking Over the Fence at Networking Jennifer Rexford.
Building a Strong Foundation for a Future Internet Jennifer Rexford Princeton University
Stable Internet Routing Without Global Coordination Jennifer Rexford AT&T Labs--Research
Multipath Protocol for Delay-Sensitive Traffic Jennifer Rexford Princeton University Joint work with Umar Javed, Martin Suchara, and Jiayue He
The Future of the Internet Jennifer Rexford ’91 Computer Science Department Princeton University
Stable Internet Routing Without Global Coordination Jennifer Rexford AT&T Labs--Research Joint work with Lixin Gao.
Rethinking Internet Traffic Management Using Optimization Theory Jennifer Rexford Princeton University Joint work with Jiayue He, Martin Suchara, Ma’ayan.
Computer Networks Layering and Routing Dina Katabi
EQ-BGP: an efficient inter- domain QoS routing protocol Andrzej Bęben Institute of Telecommunications Warsaw University of Technology,
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jennifer Rexford Princeton University With Jiayue He, Rui Zhang-Shen, Ying Li,
9/15/2015CS622 - MIRO Presentation1 Wen Xu and Jennifer Rexford Department of Computer Science Princeton University Chuck Short CS622 Dr. C. Edward Chow.
1 Interdomain Routing (BGP) By Behzad Akbari Fall 2008 These slides are based on the slides of Ion Stoica (UCB) and Shivkumar (RPI)
CS 3700 Networks and Distributed Systems Inter Domain Routing (It’s all about the Money) Revised 8/20/15.
Jennifer Rexford Fall 2014 (TTh 3:00-4:20 in CS 105) COS 561: Advanced Computer Networks BGP.
Scalable Multi-Class Traffic Management in Data Center Backbone Networks Amitabha Ghosh (UtopiaCompression) Sangtae Ha (Princeton) Edward Crabbe (Google)
A Utility-based Approach to Scheduling Multimedia Streams in P2P Systems Fang Chen Computer Science Dept. University of California, Riverside
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jiayue He, Rui Zhang-Shen, Ying Li, Cheng-Yen Lee, Jennifer Rexford, and Mung.
Intradomain Traffic Engineering By Behzad Akbari These slides are based in part upon slides of J. Rexford (Princeton university)
Stable Internet Routing Without Global Coordination Jennifer Rexford Princeton University Joint work with Lixin Gao,
Evolving Toward a Self-Managing Network Jennifer Rexford Princeton University
Jennifer Rexford Fall 2014 (TTh 3:00-4:20 in CS 105) COS 561: Advanced Computer Networks TCP.
Lowering the Barriers to Innovation Jennifer Rexford Computer Science Department Princeton University
Evolving Toward a Self-Managing Network Jennifer Rexford Princeton University
Michael Schapira, Princeton University Fall 2010 (TTh 1:30-2:50 in COS 302) COS 561: Advanced Computer Networks
Internet Traffic Engineering Motivation: –The Fish problem, congested links. –Two properties of IP routing Destination based Local optimization TE: optimizing.
Separating Routing From Routers Jennifer Rexford Princeton University
1 Internet Routing: BGP Routing Convergence Jennifer Rexford Princeton University
15-849: Hot Topics in Networking Policy and Networks Srinivasan Seshan 1.
P4P: Proactive Provider Assistance for P2P Haiyong Xie Yale University.
Introduction to Internet Routing
Can Economic Incentives Make the ‘Net Work?
COS 561: Advanced Computer Networks
COS 561: Advanced Computer Networks
COS 561: Advanced Computer Networks
Fixing the Internet: Think Locally, Impact Globally
Presentation transcript:

Building a Strong Foundation for a Future Internet Jennifer Rexford ’91 Computer Science Department (and Electrical Engineering and the Center for IT Policy)

2 Clean-Slate Network Architecture Clean-slate architecture –Without constraints of today’s artifacts –To have a stronger intellectual foundation –And move beyond the incremental fixes Still, some constraints inevitably remain –Resource limitations (CPU, memory, bandwidth) –Time delays between nodes –Independent economic entities –Malicious parties –The need to evolve over time

3 An Open Research Challenge Distributed Protocols (decentralized coordination between many hosts, routers) Autonomous Actors (autonomous parties, with different economic objectives) Global Properties (stability, scalability, reliability, security, privacy, managability, …) ?

4 Two Forays into a Theory of Networks Example topic: Internet routing –What paths should carry the traffic? –How should these paths be computed? –How much traffic should traverse each path? Two theoretical approaches –Game theory  Interdomain routing driven by economic incentives –Optimization theory  Protocol as an implicit solution to optimization problem

Game Theory: Applied to Interdomain Routing + $$$ = ???

6 What is an Internet? A “network of networks” –Networks run by different institutions Autonomous System (AS) –Routers run by a single institution –With a local routing policy ASes have different goals –Different views of which paths are good Interdomain routing reconciles those views –Computes end-to-end paths through the Internet

7 Conflicting Policies Cause Oscillation Pick the highest-ranked path consistent with your neighbors’ choices. Only choice! Top choice! Only choice! Better choice! Only choice! Better choice!

8 Economic Incentives Save the Day Two common business relationships –Customer-provider (e.g., Princeton and USLEC) –Peer-peer (e.g., AT&T and Sprint) Three economic incentives –No “transit service” for peers and providers –An AS is not its own indirect customer –Prefer routes through paying customers Provably ensures a stable system! d 4 5

9 Moving Beyond the Original Model How might business relationships evolve? –Backup routes –Siblings How might economic incentives evolve? –Desire to attract traffic from others –Accidental and malicious behavior How much can we rely on incentives? –Are we willing to rely on economic incentives to assure the most basic properties of the Internet? –Do we really have a choice?

Optimization Theory: Applied to Traffic Management

11 Traffic Management Today How much traffic should traverse each path? End hosts: Congestion control Operator: Traffic engineering Routers: Compute paths

12 Rethinking Traffic Management What should be the objective of the system? Source rate x i U i (x i ) Maximize User Utility Link utilization u l f(u l ) Minimize Link Congestion Goal: max ∑ i U i (x i ) - w∑ l f(u l )

13 Routers: Compute link “prices” Distributed Solutions Protocols as distributed optimizers –Link prices: higher as load approaches capacity –Path rates: sending traffic over cheaper paths Edge nodes: Adjust path sending rates s s s

14 Moving Beyond the Original Model Evaluating protocols in practice –Speed of convergence to optimal flow of traffic –Sensitivity to tunable parameters Considering other objectives –E.g., minimizing end-to-end delay Supporting multiple classes of traffic –Mix of throughput-sensitive and delay-sensitive –Dynamically adapting the sharing of bandwidth Deploying the protocols in real networks!

15 Conclusion Computer networks are key infrastructure –The stakes are very high –Designs that are worthy of society’s trust Inherently an interdisciplinary problem –Game theory, optimization theory, control theory, cryptography, computer science, coding theory –Domain knowledge and an experimental mindset We are making real progress –On creating a “science of computer networks” –And on building and deploying new solutions