Multiple co-existing grouping overlays

ABSTRACT

This talk discusses issues in defining multiple overlays within P2P networks for the caching of application-data as well as conventional metadata i.e. adverts, or locations to data. I will discuss work carried out at Cardiff in this area that is being built on top of the P2PS middleware, which allows new types of super-peer overlays to be defined through grouping. Such overlays extend super-peer capabilities to help programmers create schemes for defining new ways for locating data sets as well as adverts. Data-caching groups co-exist with the conventional super-peer overlay, which still bootstraps the network by providing the discovery backbone. Parameters for super-peer or data caching groups can also be adjusted e.g. number of super-peer conenctions, caching policies (replication), security policies, etc, and we are also building interfaces for defining new matchmaking algorithms (for matching adverts with queries) as well as more sophisticated schemes for organising super peers (e.g. distributed hashtables, routing tables, agents, etc.) to be plugged into the network.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.4 Distributed Systems
          Subjects: Distributed applications

General Terms:
Reliability

Keywords:
P2P, scalability, web services