Performance Study of Congestion Price Based Adaptive Service Xin Wang, Henning Schulzrinne (Columbia University) http://www.cs.columbia.edu/xinwang/public /projects/RNAP.html 1 Motivation • Combine resource reservation with multimedia adaptive service • Pricing network services based on level of service, usage, and congestion - a natural and fair incentive for applications to adapt their sending rates according to network conditions. • Our work: – resource commitment short interval – trade-offs between blocking and raising prices in network 2 Outline • Resource negotiation & RNAP • Pricing strategy • User adaptation • Simulation model • Results and discussion • RNAP message aggregation • Conclusion 3 Resource Negotiation & RNAP • Assumption: network provides a choice of delivery services to user – e.g. diff-serv, int-serv, best-effort, with different levels of QoS – with a pricing structure (may be usagesensitive) for each. • RNAP (Resource Negotiation and Pricing): a protocol through which the user and network (or two network domains) negotiate network delivery services. – Network -> User: communicate availability of services; price quotations and accumulated charges – User -> Network: request/re-negotiate specific services for user flows. • Underlying Mechanism: combine network pricing with traffic engineering 4 Resource Negotiation & RNAP (cont’d.) • Who can use RNAP? – Adaptive applications: adapt sending rate, choice of network services – Non-adaptive applications: take fixed price, or absorb price change 5 Protocol Architectures: Centralized Architecture (RNAP-C) NRN NRN NRN HRN HRN S 1 R 1 Access Domain - A Access Domain - B Transit Domain Internal Router Edge Router NRN Network Resource Negotiator HRN Host Resource Negotiator Host Data RNAP Messages Intra domain messages 6 Protocol Architectures: Distributed Architecture (RNAP-D) HRN HRN S 1 R 1 Access Domain - A Access Domain - B Transit Domain Internal Router NRN Edge Router HRN Host Data RNAP Messages Intra domain messages 7 RNAP messages Query: User enquires about available services, prices Query Quotation Reserve Periodic re-negotiation Commit Quotation: Network specifies services supported, prices Reserve: User requests service(s) for flow(s) (Flow IdService-Price triplets) Quotation Reserve Commit Close Release Commit: Network admits the service request at a specific price or denies it (Flow IdService-Status-Price) Close: tears down negotiation session Release: releases the resources 8 Pricing on Current Internet • Access rate dependent charge (AC) • Volume dependent charge (V) • AC + V AC-V • Usage based charging: time-based / volumebased 9 Pricing Strategy • Fixed pricing – Service class independent flat pricing – Service class sensitive priority pricing – Time dependent time of day pricing – Time-dependent service class sensitive priority pricing 10 Pricing Strategy (cont’d.) • Congestion-based Pricing – Usage charge: pu= f (service, demand, destination, time_of_day, ...) cu(n) = pu x V (n) – Holding charge: Phi = i x (pui - pu i-1) ch (n) = ph x R(n) x – Congestion charge: pc (n) = min [{pc (n-1) + (D, S) x (DS)/S,0 }+, pmax] cc(n) = pc(n) x V(n) 11 User Adaptation • Based on perceived value • Maximize total utility over the total cost • Constraint: budget, min QoS & max QoS 12 User Adaptation (cont’d.) • An example utility function – U (x) = U0 + log (x / xm) • Optimal user demand – Without budget constraint: xj = j / pj – With budget constraint: xj = (b x j / Σl l ) / pj • Affordable resource is distributed proportionally among applications of the system, based on the user’s preference and budget for each application. 13 Simulation Model Topology 1 Topology 2 14 Simulation Model (cont’d.) • Parameter Set-up – – – – – – – – – topology1: 48 users topology 2: 360 users user requests: 60 kb/s -- 160 kb/s targeted reservation rate: 90% price adjustment factor: σ = 0.06 price update threshold: θ = 0.05 negotiation period: 30 seconds usage price: pu = 0.23 cents/kb/min average session length 10 minutes, exponential distributed. • Performance measures – – – – – – Bottleneck bandwidth utilization User request blocking probability Average and total user benefit Network revenue System price User charge 15 Design of the Experiments • Performance comparison of congestion-based pricing system (CPA) with a fixed-price based system (FP) • Effect of system control parameters: – target reservation rate – price adjustment step – price adjustment threshold • • • • Effect of user demand elasticity Effect of session multiplexing Effect when part of users adapt Session adaptation and adaptive reservation 16 Bottleneck Utilization Request blocking probability 17 Total network revenue ($/min) Total user benefit ($/min) 18 Price ($/kb/min) User bandwidth (kb/s) 19 Average user charge ($/min) 20 Effect of target utilization level Bottleneck utilization Request blocking probability 21 Effect of Price Adjustment Step Bottleneck utilization Request blocking probability 22 Effect of Price Adjustment Threshold Bottleneck utilization Request blocking probability 23 Effect of User Demand Elasticity Average user bandwidth Average user charge 24 Effect of Multiplexing Between User Sessions Request blocking probability Total user benefit 25 Adaptation by Part of User Population Bottleneck utilization Request blocking probability 26 One-time Versus Ongoing Adaptation Bottleneck utilization Request blocking probability 27 RNAP Message Aggregation RNAP-D RNAP-C 28 RNAP Message Aggregation (cont’d) • Aggregate for senders sharing the same destination network – merged by source domains – split for HRNs at destination net: border router (RNAP-D) NRN (RNAP-C) • Two messages: – aggregated-resource message reserves and collects price in the middle of network – original messages sent directly to destination without visiting agents in between 29 Conclusions • CPA gain over FP – Network availability, revenue, user perceived benefit – CPA congestion control is stable and effective • Target reservation rate (utilization): – too high or too low utilization: User benefit – Too low target rate: demand fluctuation is high – Too high target rate: high blocking rate 30 Conclusions • Effect of price scaling factor – , blocking rate – too large under-utilization, large dynamics • Effect of price adjustment threshold – Too high, no meaningful adaptation 31 Conclusions • Demand elasticity – Bandwidth sharing is proportional to user’s willingness to pay • User adaptation by some users still results in overall performance improvement 32