html5-img
1 / 29

Edge-to-Edge Bailout Forward Contracts for Single-Domain Internet Services

Edge-to-Edge Bailout Forward Contracts for Single-Domain Internet Services. Weini Liu , Rensselaer Polytechnic Institute Hasan Karaoglu , University of Nevada, Reno Aparna Gupta , Rensselaer Polytechnic Institute Murat Yuksel , University of Nevada, Reno

hanh
Download Presentation

Edge-to-Edge Bailout Forward Contracts for Single-Domain Internet Services

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Edge-to-Edge Bailout Forward Contracts for Single-Domain Internet Services Weini Liu, Rensselaer Polytechnic Institute Hasan Karaoglu, University of Nevada, Reno Aparna Gupta, Rensselaer Polytechnic Institute Murat Yuksel, University of Nevada, Reno Koushik Kar, Rensselaer Polytechnic Institute

  2. Motivation • ISPs can provide single domain QoS for wide range of applications • Video-conference • Online Video Games • Streaming Live Media, name it

  3. Motivation • There is no easy way to : • express QoS demand to intermediate providers • exchange QoS service capabilities and offers to other providers ?

  4. Motivation Implied Challenges • Current architectural problems: • Providersdo not have economic knobs to manage risks involved in • investing innovative QoS technologies and • business relationships with other providers • Userscannot express value choices at sufficient granularity – only at access level capability to provide e2e higher quality services money-back guarantees, risk/cost sharing flexibility in time: forward/option pricing flexibility in space: user-defined inter-domain routes

  5. Stations of the provider computing and advertising local prices for edge-to-edge contracts. Stations of the provider computing and advertising local prices for edge-to-edge contracts. Edge Router Edge Router Edge Router Customers Network Core Edge Router Edge Router Edge Router Intra-domain Dynamic Contracts • Contract components • performance component, e.g., capacity • financial component, e.g., price • time component, e.g., term

  6. Contract Switching Paradigm

  7. ISP B ISP A ISP C ISP B routable datagrams ISP A ISP C ISP B contracts overlaid on routable datagrams ISP A ISP C Contract-switching: A paradigm shift… Circuit-switching e2e circuits Packet-switching Contract-switching 7

  8. SPOT, Forward and Bailout Forward Contracts • Well defined components (performance, financial and time) • Time Duration for Contracts • Spot Contracts • Providing the service in short term considering current market conditions • Forward Contracts • Providing the service at a future time for predetermined price • Bailout Forward Contracts • Make forward contracts applicable to capacitated network resources Atomic Short Medium Long

  9. Bailout Forward Contract • Single g2g Contract • Bandwidth guarantee as performance metric ( eg.100 Mbps ) • Bandwidth threshold to activate bailout condition ( e.g..80 Mbps) • Exposed to variation on demand, bandwidth capacity and spot prices • (e.g. For next five days, $0.25)

  10. Bailout Forward Contract • Multiple g2g Contracts • Exposed to the additional effects of other traffic flows as a result of overlapping links with other g2g overlay paths • How we model this interaction ?

  11. Intensity of Overlap Link Capacity 1.5 gbps Utilization = 10% Link Capacity 0.3 gbps Utilization = 50% Flow1 50 mbps Flow2 100 mbps

  12. Performance • Can an ISP survive by applying BFC approach ? • How frequent does BFC bailout ? • BFC Robustness against • increasing demand ? (demand) • decreasing available bandwidth ? (supply) • major link failures ? • How efficient is BFC pricing ? • Revenue Losses

  13. Network Model • Realistic Simulation requires • Realistic ISP Topology • Adjacency Matrix (Given by Rocketfuel Data)‏ • Link Delays & Weights (Given by Rocketfuel Data)‏ • Link Capacities (We have to model)‏ • Edge and Backbone Router Classification (We have to model)‏ • Routing Matrix (Path calculated by Shortest Path Algorithms, as OSPF does)‏ • Realistic Traffic Model • Traffic Matrix Estimation(We have to model)‏

  14. Network Model • BFS Based Link Capacity Estimation • Select Most Connected Router • Initiate a BFS • Assign higher capacity to links closer to center DC. BFS Dist = 2 Atlanta BFS Dist = 3

  15. Network Model Gravity Model Traffic ~ Pop. 1 X Pop. 2 Seattle 6K 24X 2X Chicago 3M NY 8M • Gravity Model Based Traffic Estimation • Determine Degree and BFS Distance Thresholds • Classify Edge / Backbone Routers • Associate Area Population with Edge Router s • Use Gravity Model to estimate traffic size

  16. Network Model Summary : • Construct Routing Matrix RnXn • Using Adjacency Matrix A and Weight Matrix W given by Rocketfuel Topologies • Apply Dijktra’s Shortest Path Algorithm • BFS Based Link Capacity Estimation • Form the Traffic Matrix TnXnusing Gravity Model • Calculate traffic load on each link • If link capacity is less than traffic load, upgrade link capacity

  17. Simulation Setup – Multiple BFC’s • Failure Mode 1-2-3 • One of the most heavily loaded link is failed • Traffic passes through failed link is rerouted • Bandwidth share and correlation between traffic flows also changes

  18. Simulation Results – Single BFC

  19. Simulation Results– Single BFC Results for five sample g2g contracts Threshold = 15th percentile

  20. Simulation Results– Multiple BFCs Mean fraction 16.4 %

  21. Simulation Results– Multiple BFCs Bailout times intensify around 150 / 1000 ~ 15% level

  22. Simulation Results– Multiple BFCs Bailout times intensify around 150 / 1000 ~ 15% level

  23. Simulation Results– Multiple BFCs Bailout times intensify around 150 / 1000 ~ 15% level

  24. Simulation Results– Multiple BFCs Bailout times intensify around 150 / 1000 ~ 15% level

  25. Simulation Results– Multiple BFCs

  26. Network Analysis Mean fraction 27%

  27. Summary • Single domain edge to edge path abstraction • Domain is a set of G2G overlay paths • Flexibility to price each G2G path individually • Bailout Forward Contract with Bandwidth Guarantees • Bandwidth threshold as Bailout clause • Money-back guarantees • BFC’s are robust against link failures, fluctuations in demand, supply and correlation effects between traffic flows

  28. Future Work • Conduct network analysis inter-domain e2e QoS paths “Contract Paths” which are composed of edge to edge (g2g) intra-domain “Contract Links” • Study Economical and Technical Aspects of “Contract Switching Paradigm” • Discovery of Better Alternative Paths • Better network utilization • Game Theoretical analysis of collaboration of ISPs

  29. Questions ? • Value Flows and Risk Management Architecture for Future Internet • Project Website • http://www.nets-find.net/Funded/ValueFlows.php • http://www.cse.unr.edu/~yuksem/contract-switching.htm • Or google “Contract Switching”

More Related