1 / 18

REPS: Scalable Reputation Management for P2P MMOGs

REPS: Scalable Reputation Management for P2P MMOGs. Guan-Yu Huang, Shun-Yun Hu, Jehn-Ruey Jiang National Central University, Taiwan MMVE’08, Reno, USA, 2008 2008/03/08. Introduction Problem formulation and challenges Design of REPS Discussions Conclusion. Outline.

candid
Download Presentation

REPS: Scalable Reputation Management for P2P MMOGs

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. REPS: Scalable Reputation Management for P2P MMOGs Guan-Yu Huang, Shun-Yun Hu, Jehn-Ruey Jiang National Central University, Taiwan MMVE’08, Reno, USA, 2008 2008/03/08

  2. Introduction • Problem formulation and challenges • Design of REPS • Discussions • Conclusion Outline

  3. Booming of Virtual Environments (VEs) • Massively Multiplayer Online Games (MMOGs) Virtual Environments

  4. Architectures: • Client-server • All loads are on the server • Peer-to-Peer (P2P) • Distribute loads to all users • More scalable & affordable • Based on locality of interactions P2P-based MMOGs

  5. MMOG requirements • Consistency Responsiveness • Persistency Reliability • Security Scalability • Security issues • Data modification • Game rules mis-processing Problems in P2P-MMOGs

  6. Reputation may identify trustworthy users • Reputation is feasible • MMOGs are socially-oriented • Players value in-game status • Reputation is useful • To decide whether to interact • To delegate responsibilities • REPS for P2P MMOGs • Localized trust evaluation with rating right • Selection oftrust nodes to store & query reputations Motivation & Proposed method

  7. How to store reputation scores on reliable peers and query them effectively? • Assumptions • Fixed AOI radius • P2P-VE overlay provides AOI neighbors • Users may mutually rate each other Problem formulation

  8. Reputation evaluation • Precise • Simple • Storage and query • Scalable • Efficient • Reliability • Cheat-proof • Failure-proof Challenges

  9. Mutual rating • AOI-only • Rating Right • Given only after interactions within AOI • Rate once, modify later any time • Positive (1), Negative (-1), Neutral (0) Local reputation evaluation

  10. Trust nodes to store reputation values • Chosen from AOI neighbors (may time-out) • List of trust nodes stored as trust list at each user • Storage • Obtain trust list • Send evaluations to trust nodes directly • Query • Obtain trust list • Randomly choose n trust nodes (out of N total) • Majority decision Reputation storage and query

  11. Combines two intuitive factors • Total score: TS (u) • Total rating: V (u) • Positive ratio: P (u) • Which metric is more important for selecting a given number of trust nodes? Neighbor Trust node Selection (NTS)

  12. Positive Ratio (P) High Low High Trust Region Total Score (TS) Trust nodes Other nodes Regression Line Regression line Low NTS (trust region)

  13. Positive Ratio (P) High Low High Total Score (TS) New trust node Trust nodes Other nodes Regression line Low NTS (adjustment along positive ratio)

  14. Positive Ratio (P) High Low High Total Score (TS) New trust node Trust nodes Other nodes Regression line Low NTS (adjustment along total score)

  15. Reputation evaluation • Simple representation for reputations • Rating right control • Storage and query • Remote storage prevents self-modification • Distribution enhances scalability • Reliability • Socially enforced mutual monitoring • Majority-based value retrieval masks cheating Discussions

  16. Not 100% secure (but may converge) • Incentives for rating (need actual tests) • Query efficiency (due to replications) • Bootstrapping (will converge?) Issues

  17. Reputation management for P2P MMOGs • Mutual rating • Distributed storage (trust nodes) • Characteristics • Low server overhead • Consensus-based monitoring • Cheat-proof measures Conclusion

  18. Thank You Q & A

More Related