1 / 28

Incentive Mechanisms for Large Collaborative Resource Sharing

Objectives: Why Resource harnessing Examples of resource harnessing Grid computing P2P computing Resource sharing Assumptions Considerations What are incentives? Trust as a mechanism to provide incentives. Incentive Mechanisms for Large Collaborative Resource Sharing.

curry
Download Presentation

Incentive Mechanisms for Large Collaborative Resource Sharing

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. Objectives: Why Resource harnessing Examples of resource harnessing Grid computing P2P computing Resource sharing Assumptions Considerations What are incentives? Trust as a mechanism to provide incentives Incentive Mechanisms for Large Collaborative Resource Sharing

  2. Resource Harnessing • Huge interest in linking up resources • Grid computing, P2P computing, computing utilities, etc. • It is all about sharing • Quality of Service • Security • Participation versus Cost

  3. Resource Harnessing: Grid Example • Virtual Private Grids (PVG) is a framework for “renting” collection of resources • “Collection” is defined as follows: • able to deliver predefined performance metrics • performance delivered at predefined geographical locations • cost of provisioning is optimized or bounded

  4. Grid Resource base Resource Harnessing: Grid Example VPGR GR GR GR multiplex Grid Domain Grid Resource Grid Resource Grid Resource

  5. Resource Harnessing: Grid Example SO • SO (service originator) presents the VPG Spec. via a VPG Manager (VPGM) • VPGM negotiates with different Grids via a MetaGrid Resolver (MGR) • Grids (GRs) bid for the VPG creation requests • VPGM selects the best bid • Location spec • QoS specs • Cost preference VPGS VPGM Contract negotiation Admission Control MGR bid with (QoS/cost) VPG creation request Grid Engineering GR GR …… GR

  6. Resource Sharing • Assumptions • Resource owners have committed their resources • Honestly • To be used efficiently • To be used for the overall good of the community • Considerations • Free riding • Malicious entities • Non cooperative entities Incentives are needed for resources to cooperate honestly

  7. Resource Harnessing: P2P Example • Since, we deal with public resources, we need to address the following • How can we encourage resources to cooperate • 70% of all users do not share files • 50% of all requests are satisfied by the top 1% sharing hosts • How can we deal with security • We do not want security to become an overhead! • Can we use “trust” as an incentive?

  8. Trust Considerations • How can we define “trust” in an operational way? Who will evaluate trust? • Trust maintenance can result in an efficient process especially in a very large-scale system. Hence, our task is to come up with an efficient model for maintaining trust • Techniques for managing and evolving trust in a large-scale distributed system • Mechanisms for maintaining trust from ongoing transactions

  9. Overall Trust Model

  10. Trust Terminology • Identity trust • Behavior trust • Honesty • Accuracy • Set of recommenders • Set of trusted allies

  11. Trust Model Characteristics • To make the trust model efficient • the overall NC system is divided into NCDs • trust is a slow varying attribute • the number of contexts is limited to printing, storage, and computing

  12. Why Behavior Trust

  13. Notation • Let and represent recommenders set and trusted allies set, respectively • Let the honesty of recommender as observed by be denoted as • Let denote the recommendation for given by to at time for context • Let denote the recommendation for given by to where for the same and

  14. Computing Honesty • Let • The value of will be less than a small value if recommender is honest • Therefore, is computed as

  15. Computing Accuracy • Let denote the true trust level of obtained by as a results of monitoring the transaction • Let • The value of will be an integer value ranging from 0 to 4 • Therefore, is computed as

  16. Computing Trust & Reputation • Before can use the recommendation given by to calculate the reputation of , needs to be adjusted to reflect the accuracy of recommender • This shift is given by

  17. Computing Trust & Reputation • Trust relationship expressed as • Direct trust relationship and the reputation of expressed as and ,respectively. • The decay function is expressed as • Let and

  18. Simulation Setup • A discrete event simulator was used • The transactions arrival process modeled using a Poisson random process • 30 NCDs were used in the simulation • The size of R is fixed and set to 4 • The size of T is fixed and set to 3 • The TL were randomly generated from [1-5]

  19. Performance Measurement • The measure of performance used is the ability of the trust model to correctly predict the trust that exists between two NCDs • This is quantified by determining the success ratio as follows:

  20. Performance Evaluation • Using accuracy & honesty measures: Success ratio with 150 transactions per relation

  21. Performance Evaluation • Using the accuracy measure: Success ratio with 150 transactions per relation

  22. Performance Evaluation • Using Accuracy & honesty measures: Success ratio progress

  23. Case Study: Trust Modeling on P2P Grids • The P2P Grid is segmented into Grid domains (GDs) • Two virtual domains are associated with each GD • resource domain and client domain • Each resource domain has 3 attributes: • Ownership • Type of Activities (ToA) it supports • TL for each ToA • Similarly, each client domain has 3 attributes

  24. Case Study: Trust Modeling on P2P Grids • Suppose that client from wanting to engage in activities and on resource at • Offered TL (OTL) = min(TL for , TL for ) • There are two required TLS (RTLs) • one from the client domain • one from the resource domain • Expected trust supplement (ETS) = RTL - OTL

  25. Case Study: Trust Modeling on P2P Grids • An example of the ETS table

  26. Case Study: Trust Modeling on P2P Grids • A batch mode mapping heuristic called “Sufferage heuristic” was used

  27. Case Study: Trust Modeling on P2P Grids • Two different classes of Expected Execution Cost (EEC) were used: • Consistent Low task low machine (LOLO) heterogeneity • models networks that have “related” machines which are “similar” in performance • Inconsistent Low task low machine (LOLO) heterogeneity • models networks were machines are not related

  28. Case Study: Performance Evaluation

More Related