Rade Stanojevic Using Tuángòu to Reduce IP Transit Costs Rade Stanojevic (Joint work with Ignacio Castro and Sergey Gorinsky) ‍‍ ACM CoNext 2011 Tokyo,

Slides:



Advertisements
Similar presentations
Achieving Price-Responsive Demand in New England Henry Yoshimura Director, Demand Resource Strategy ISO New England National Town Meeting on Demand Response.
Advertisements

Vytautas Valancius, Cristian Lumezanu, Nick Feamster, Ramesh Johari, and Vijay V. Vazirani.
A Short Tutorial on Cooperative Games
Fast Algorithms For Hierarchical Range Histogram Constructions
Market-Based Management
Chapter 10 Basics of Saving and Investing
Employing Agent-based Models to study Interdomain Network Formation, Dynamics & Economics Aemen Lodhi (Georgia Tech) 1 Workshop on Internet Topology &
1/20 On the Shapley-like Payoff Mechanisms in Peer-Assisted Services with Multiple Content Providers April 17, 2011 JEONG-WOO CHO KAIST, South Korea Joint.
1/27 Evaluating Potential Routing Diversity for Internet Failure Recovery *Chengchen Hu, + Kai Chen, + Yan Chen, *Bin Liu *Tsinghua University, + Northwestern.
1 Analysis of Topographical Leverage- driven Capacity Trading in Internet Storage Infrastructures Anna Ye Du (SUNY, Buffalo), Xianjun Geng (UW, Seattle),
Introduction to Exchange Point Economics Version 3.0 January, 2002 Bill Woodcock Packet Clearing House.
SM0374 Strategic Management and Leadership Lecture 7: Strategic Capabilities 3.
Understanding the Accuracy of Assembly Variation Analysis Methods ADCATS 2000 Robert Cvetko June 2000.
Probabilistic Aggregation in Distributed Networks Ling Huang, Ben Zhao, Anthony Joseph and John Kubiatowicz {hling, ravenben, adj,
The Allocation of Value For Jointly Provided Services By P. Linhart, R. Radner, K. G. Ramkrishnan, R. Steinberg Telecommunication Systems, Vol. 4, 1995.
Agent-based Modeling: Methods and Techniques for Simulating Human Systems Eric Bonabaun (2002) Proc. National Academy of Sciences, 99 Presenter: Jie Meng.
Internet Research Needs a Critical Perspective Towards Models –Sally Floyd –IMA Workshop, January 2004.
Adhva Finding Customers and Forecasting Markets for New and Disruptive Technologies James F. Fee, Adhva Adhva.com.
11 Quantifying Benefits of Traffic Information Provision under Stochastic Demand and Capacity Conditions: A Multi-day Traffic Equilibrium Approach Mingxin.
Approximating Power Indices Yoram Bachrach(Hebew University) Evangelos Markakis(CWI) Ariel D. Procaccia (Hebrew University) Jeffrey S. Rosenschein (Hebrew.
Market-Based Management
THE ROLE OF PROACTIVE ADAPTATION IN INTERNATIONAL MITIGATION COALITIONS Kelly de CERE.
Tradeoffs in CDN Designs for Throughput Oriented Traffic Minlan Yu University of Southern California 1 Joint work with Wenjie Jiang, Haoyuan Li, and Ion.
Internet Inter-Domain Traffic Craig Labovitz, Scott Iekel-Johnson, Danny McPherson, Jon Oberheide, Farnam Jahanian Presented by: Kaushik Choudhary.
Ensemble Learning (2), Tree and Forest
Network Management and Internet Pricing S. Raghavan J. Bailey The Robert H. Smith School of Business University of Maryland.
Internet Infrastructure and Pricing. Internet Pipelines Technology of the internet enables ecommerce –Issues of congestion and peak-load pricing –Convergence.
© GSM Association 2011 Mobile Energy Efficiency Mobile and Environmental Sustainability 20 September 2012.
1 Patch Complexity, Finite Pixel Correlations and Optimal Denoising Anat Levin, Boaz Nadler, Fredo Durand and Bill Freeman Weizmann Institute, MIT CSAIL.
Including a detailed description of the Colorado Growth Model 1.
© XchangePoint 2001 Economic Differences Between Transit and Peering Exchanges Keith Mitchell Chief Technical Officer NANOG 25 10th June 2002.
Chapter 26 Pricing Strategies.
Impact of Prefix Hijacking on Payments of Providers Pradeep Bangera and Sergey Gorinsky Institute IMDEA Networks, Madrid, Spain Developing the Science.
MBA7020_04.ppt/June 120, 2005/Page 1 Georgia State University - Confidential MBA 7020 Business Analysis Foundations Descriptive Statistics June 20, 2005.
NAB 2012 Cloud Computing Conference New Levels of Media Performance Data Enabled by Cloud Computing -- and Impact on Other Sectors Scott Brown, GM US and.
Milk, Feed, and MPP Margin Price Forecasting John Newton University of Chuck Nicholson Penn State University.
Aditya Akella The Performance Benefits of Multihoming Aditya Akella CMU With Bruce Maggs, Srini Seshan, Anees Shaikh and Ramesh Sitaraman.
Information Systems for Competitive Advantage Source: Management Information System, 10 edition Raymond McLeod & George Schell.
Valuation of Asian Option Qi An Jingjing Guo. CONTENT Asian option Pricing Monte Carlo simulation Conclusion.
TOMA: A Viable Solution for Large- Scale Multicast Service Support Li Lao, Jun-Hong Cui, and Mario Gerla UCLA and University of Connecticut Networking.
Aemen Lodhi (Georgia Tech) Amogh Dhamdhere (CAIDA)
How AWS Pricing Works Jinesh Varia Technology Evangelist.
PPT Characteristics of Good and Poor Advertisements PPT
Measuring the Effect of Queues on Customer Purchases Andrés Musalem Duke University Joint work with Marcelo Olivares, Yina Lu (Decisions Risk and Operations,
4. Numerical Integration. Standard Quadrature We can find numerical value of a definite integral by the definition: where points x i are uniformly spaced.
ITGS Databases.
Can ISPs be Profitable Without Violating Network Neutrality? Amogh Dhamdhere Constantine Dovrolis Georgia Tech.
Lecture PowerPoint Slides Basic Practice of Statistics 7 th Edition.
Customer Service & Logistics Filmco & Walmart. Components of Customer Service Before Pre-Transaction Service During Transaction Service After Post-Transaction.
K-20 Co-pay and ISP Basics Valid for Fiscal Years 2010 & 2011 (July 1, 2009 – June 30, 2011)
Marc IVALDI Workshop on Advances on Discrete Choice Models in the honor of Daniel McFadden Cergy-Pontoise – December 18, 2015 A Welfare Assessment of Revenue.
ApproxHadoop Bringing Approximations to MapReduce Frameworks
CEEDS Phase 3: Energy Saving Potential in Urban Passenger Transportation 5 March, 2012 Ralph D. Samuelson APERC Workshop, Kuala Lumpur 1.
1/18 Evaluating Potential Routing Diversity for Internet Failure Recovery *Chengchen Hu, + Kai Chen, + Yan Chen, *Bin Liu *Tsinghua University, + Northwestern.
© 2012 Cengage Learning. All Rights Reserved. Principles of Business, 8e C H A P T E R 10 SLIDE Marketing Basics Develop Effective.
Peering and Interconnection Economics Introduction to Internet Transit and Peering.
SCREAM: Sketch Resource Allocation for Software-defined Measurement Masoud Moshref, Minlan Yu, Ramesh Govindan, Amin Vahdat (CoNEXT’15)
© 2012 Cengage Learning. All Rights Reserved. May not scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. Chapter.
Calculating Prices Bait-and-switch advertising: Promoting a low-priced item to attract customers to whom the business then tries to sell a higher.
NEW PRODUCT LICENSING: Why do companies license their IP?
Introduction to Exchange Point Economics
CS 4700 / CS 5700 Network Fundamentals
Microeconomics: Chapter 1
Amogh Dhamdhere (CAIDA/UCSD) Constantine Dovrolis (Georgia Tech)
Internet Interconnection
Internet Traffic Management: Security vs. Economics
Confidence Intervals: The Basics
Ensemble learning Reminder - Bagging of Trees Random Forest
Internet Exchange.
Robert C. Chalmers and Kevin C. Almeroth
Presentation transcript:

Rade Stanojevic Using Tuángòu to Reduce IP Transit Costs Rade Stanojevic (Joint work with Ignacio Castro and Sergey Gorinsky) ‍‍ ACM CoNext 2011 Tokyo, Japan 团购 (tuángòu) = group-buying

Rade Stanojevic Part 1: Tuángòu for IP transit ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs2

Rade Stanojevic IP transit as a service ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs3 IP transit Lower tier ISPs customers of higher tier ISPs Critical for global connectivity Fee depends on traffic

Rade Stanojevic IP Transit costs ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs4 Affects almost every online entity IP transit cost-reduction technologies ISP peering CDN Multicast P2P localization Traffic shaping (P2P throttling) The less traffic on transit links the lower the cost

Rade Stanojevic Simple idea: Exploit the IP transit market properties: Price elasticity 95 th -percentile pricing CIPT: Cooperative IP transit ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs5 purchase in group and share the costs

Rade Stanojevic IP transit billing: two key properties ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs6 Price elasticity Peak based traffic metering: 95 th -percentile pricing Submodularity of the peak: peak(f) + peak(g) ≥ peak(f+g) Voxel pricing Source= (as accessed on June 2011)

Rade Stanojevic CIPT as a cooperative game ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs7 How to share the benefits of cooperation? Shapley value ‘Fair’, exists and unique for any cooperative game Individually rational (everyone better off in the coalition) Hard to calculate exactly for large number of players Monte-Carlo method for accurate estimation

Rade Stanojevic Part 2: Quantifying CIPT benefits ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs8

Rade Stanojevic From peering to transit traffic ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs9 ISP Transit traffic data: not available ISP Peering traffic data: available at some IXPs Peering traffic as a proxy for estimating transit traffic T transit (t) = cT peering (t) Validation using 4 medium-sized (academic) ISPs

Rade Stanojevic Obtaining peering traffic data ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs10 2. Transform images into numeric data on peering traffic 1. Peering traffic data of 264 ISPs from 6 IXPs Optical function/ character recognition Collection of mrtg images

Rade Stanojevic Expected cost savings in the range of 8%-56% ISP traffic volume distribution, peak-hour profile, nmb of transit providers, etc. Per ISP savings from $1K to few $100K per year Smallest ISPs save >90%, Largest ISPs save <10% Breakdown Around 1/3 of the savings come from the 95 th -pct The rest comes from the elastic pricing Check the paper for details… Country-wide coalitions ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs11

Rade Stanojevic Small (random) coalitions provide most of total attainable gains CIPT coalition size: low of diminishing returns ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs12

Rade Stanojevic Beyond gains sharing ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs13 Organizational embodiment Physical infrastructure Performance/SLAs Traffic confidentiality Transit providers and strategic issues

Rade Stanojevic Open problems CIPT coalition formation process? Shapley value too implicit. Simpler metrics? Tuángòu in other (price-elastic) domains? IP transport Wired/wireless access Mobile voice/txt/data ACM CoNEXT CIPT: Using Tuángòu to Reduce IP Transit Costs

Rade Stanojevic Team-buying has potential for cutting IP transit bill Shapley value as a mechanism for cost sharing Unique dataset of traffic data from 264 ISPs with heterogeneous sizes and temporal properties Conclusions ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs15

Rade Stanojevic Backup slides ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs16

Rade Stanojevic Cooperative IP Transit (CIPT) ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs17 Tuángòu = team-buying coalitional arrangement for bulk buying of IP transit CIPT gains Per-Mbps price reduction thanks to elastic pricing and peak-based (95 th -pct) metering

Rade Stanojevic Price decay vs. traffic growth ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs18 Source= Per-Mbps transit price decline vs. interdomain traffic growth

Rade Stanojevic Shapley value definition ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs19 Shapley value(i) ISP i's expected marginal contribution if the players join the coalition one at a time, in a uniformly random order N = number of players c(S)= cost of coalition S S(π, i) = set of players arrived in the system not later than i π = permutations of the set of players N i’s marginal contribution

Rade Stanojevic Monte Carlo method* We estimate the Shapley value as the average cost contribution over set π k of K randomly sampled arrival orders. Estimation accuracy K is the knob controlling the accuracy We use K = 1000 to keep the error under 1% (*) D. Liben-Nowell, A. Sharp, T. Wexler, K. Woods, “Computing Shapley Value in Cooperative Supermodular Games”, Preprint, R. Stanojevic, N. Laoutaris, P. Rodriguez, “On Economic Heavy Hitters: Shapley Value Analysis of the 95th-Percentile Pricing”, Proc. of ACM IMC Shapley value estimation ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs20

Rade Stanojevic Peering = 35-40% of the total traffic. Peering and transit follow very similar patterns. Peering-transit traffic similarity ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs21

Rade Stanojevic Absolute aggregate CIPT gains grow with IXP size (in terms of billed traffic) Relative aggregate CIPT gains decrease with IXP size Aggregate CIPT gains ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs22

Rade Stanojevic Per-partner CIPT gains ACM CoNEXT 2011CIPT: Using Tuángòu to Reduce IP Transit Costs23 Absolute individual CIPT gains grow with ISP size Relative individual CIPT gains decrease with ISP size There are large gains for all CIPT members