Vytautas Valancius Cristian Lumezanu Nick Feamster Ramesh Johari
- Slides: 24
Vytautas Valancius, Cristian Lumezanu, Nick Feamster, Ramesh Johari, and Vijay V. Vazirani How Many Tiers? Pricing in the Internet Transit Market
Internet Transit Market �Sellers Large ISPs National or international reach �Buyers Cogent Traffic Invoice Smaller ISPs Enterprises Content providers Stanford University Universities Connectivity is sold at bulk using blended rates 2
What is Blended Rate Pricing? � Single price in $/Mbps/month � Charged each month on aggregate throughput Some flows are costly EU Cost: $$$ Cogent US Cost: $ Some are cheaper to serve Price is set to recover total costs + margin � Convenient for ISPs and clients Can be inefficient! Blended rate Price: $$ Stanford University 3
Issues With Blended Rate Pricing Uniform price yet diverse resource costs Clients Lack of incentives to conserve resources to costly destinations ISPs Lack of incentives to invest in resources to costly destinations � Pareto inefficient resource allocation A well studied concept in economics � Potential loss to ISP profit and client surplus Alternative: Tiered Pricing 4
Tiered Pricing Price the flows based on cost and demand �Some industries use tiered pricing extensively Parcel services, airlines, train companies Pricing on distance, weight, quality of service �Other industries offer limited tiered pricing USPS mail, London’s Tube, Atlanta’s MARTA Limited number of pricing tiers Where is tiered pricing in the Internet? 5
Tiered Pricing in The Internet Some ISPs already use limited tiered pricing Regional pricing On/Off-Net Pricing Global, Cost: $$$ Cogent Client Revenue: $ Peer No revenue Local Cost: $ Cogent Price: $$$ Stanford University Price: $$$ Stanford University Question: How efficient are the current ISP pricing strategies? Can ISPs benefit from more tiers? 6
Challenges How can we test the effects of tiered pricing on ISP profits? 1. Demand of different flows Servicing costs of different flows Modeling Data mapping Number crunching Construct an ISP profit model that accounts for: 2. Drive the model with real data Demand functions from real traffic data Servicing costs from real topology data 3. Test the effects of tiered pricing! 7
ISP Profit Model: Assumptions Profit = Revenue – Costs (for all flows) �Flow revenue Price * Traffic Demand is a function of price How do we model and discover demand functions? �Flow cost Servicing Cost * Traffic Demand Servicing Cost is a function of distance How do we model and discover servicing costs? 8
Approach to Modeling Traffic Demands Current Prices Network Topologies Demand Models Cost Models Demand Functions Relative costs 1. Finding Demand Functions Profit Model 2. Modeling Costs Absolute costs 3. Reconciling cost with demand 9
Finding Demand Functions Canonical commodity demand function: Demand = F(Price, Valuation, Elasticity) Price Inelastic demand Elastic demand Valuation – how valuable flow is Elasticity – how fast demand changes with price Demand How do we find the demand function parameters? Valuation = F-1(Price, Demand, Elasticity) Assumed range of elasticities Current price Current flow demand We mapped traffic data to demand functions! 10
Approach to Modeling Traffic Demands Current Prices Network Topologies Demand Models Cost Models Demand Functions Relative costs 1. Finding Demand Functions Profit Model 2. Modeling Costs Absolute costs 11
Modeling Costs How can we model flow costs? Linear: Concave: Region: Dest. type: ISP topologies and peering information alone can only provide us with relative flow servicing costs. real_costs = γ * relative_costs 12
Approach to Modeling Traffic Demands Current Prices Network Topologies Demand Models Cost Models Demand Functions Relative costs 1. Finding Demand Functions Profit Model 2. Modeling Costs Absolute costs 3. Reconciling cost with demand 13
Normalizing Costs and Demands Assuming ISP is rational and profit maximizing: Profit = Revenue – Costs = F(price, valuations, elasticities, real_costs) F’(price*, valuations, elasticities, real_costs) = 0 F’ (price*, valuations, elasticities, γ * relative_costs) = 0 γ = F’-1(price*, valuations, elasticities, relative_costs) Data mapping is complete: we know demands and costs! Subject to the noise that is inherent in any structural estimation. 14
Testing ISP Pricing Strategies Select a number of pricing tiers to test 1. Map flows into pricing tiers 2. 3. 1, 2, 3, etc. Optimal mapping and mapping heuristics Find profit maximizing price for each pricing tier and compute the profit Repeat above for: - 2 x demand models - 4 x cost models - 3 x network topologies and traffic matrices 15
Profit Capture Results Constant elasticity demand with linear cost model Tier 1: Local traffic Tier 2: The rest of the traffic *Elasticity – 1. 1, base cost – 20%, seed price - $20 16
Traffic and Topology Data Net. Flow records and geo-location information Group flows in to distance buckets Data Set Traffic (TB/day) Local Traffic Bit-Weighted Distance Average (miles) CV CDN 1037 ~30% 1988 0. 59 EU ISP 400 ~40% 54 0. 70 Abilene 43 ~40% 660 0. 54 Approximate measure of flow servicing cost spread 17
Results: Big Picture Linear Cost Model Concave Cost Model Constant Elasticity Demand Logit Demand 18
Future Work � Refine demand cost modeling Hybrid demand cost models are likely more realistic � Establish better metrics that predict the benefit of tiered pricing � Establish formal conditions under which demand cost normalization frameworks E. g. , can we normalize cost and demand if cost is a product of the unit cost and the log of the demand? � Test the framework on other industries 19
Summary � ISPs today predominantly use blended rate pricing � Some ISPs started using limited tiered pricing � Our study shows that having more than 2 -3 pricing tiers adds only marginal benefit to the ISP � The results hold for wide range of scenarios Different demand cost models Different network topologies and demands Large range of input parameters Questions? http: //valas. gtnoise. net 20
21
What About Competition? �Very hard to model! �Perhaps requires game-theoretic approach and more data (such as where the topologies overlap, etc. ) �It is possible to model some effects of competition by treating demand functions as representing residual instead of inherent demand. See Perloff’s “Microeconomics” pages 243 -246 for discussion about residual demand. 22
23
Caveats �We don’t know elasticities, so we test large range of them. �The data might be biased already for the traffic because of congestion signalling (maybe real demand is more than we can see). �We can’t model competition effects in long term (in fact, no one can. ) 24
- Vytautas valancius
- Vytautas dumčius
- Vytautas martinaitis
- Vytautas andrius graičiūnas
- Vytautas jurgaitis
- Vytautas byla
- Mandatory take over
- Liudna pasaka trumpas aprasymas
- Ramesh sitaraman
- Cathvideo
- Final tooth preparation stage
- Ramesh prabhu deemed conveyance
- Ramesh kaul
- Ramesh govindan usc
- Bradford vts akt
- Ramesh mehay
- Ramesh sitaraman
- Ramesh adhikari class 12
- Ramesh r. rao
- Ramesh mehay
- Ramesh mehay
- Ramesh sitaraman
- Ramesh sitaraman
- Ramesh mehay
- Berkeley algorithm example