back to main page

      1. back to main page
  2. About Distributed Search and Storage
      1. Deliverables:
      2. Architecture
      3. Components
    2. How to help
  3. Contact

About Distributed Search and Storage

The social dimension of the Nepomuk project (The Social Semantic Desktop) introduces the requirement for rich user interactions in several aspects. Therefore, it is mandatory to provide the ability of searching for users and remote resources. This subproject aims at providing components, which leverage distributed search and storage in a decentralized and heterogeneous network environment. The basic idea behind such a system is to create references to resources a user wants to share, called index entries, and distribute them into a distributed index.

Deliverables:

  • D4.1 Distributed Search System - Basic Infrastructure: Dec 2006 (EPFL, L3S) - submitted and accepted
  • D4.2 Advanced search and basic distributed storage: in progress
  • D4.3 Distributed search and storage system: planned for next year

Architecture

(A figure here)

Components

  • Distributed Index (DI) is the identified component in the Nepomuk architecture for performing this task. The implemented component is based on P-Grid, a highly scalable structured overlay network. Currently, DI supports lookup operation for exact keys, however more advanced operations are under development (e.g., range queries). Moreover, functionality for distributed storage is under development, too.
  • Distributed Storage is currently under development. Best effort replication of resources is the targeted policy.

How to help

  • End users can help by...
    • join the mailinglist, report bugs, make suggestions
    • edit these wiki pages :-)
  • Developers can help by...
    • doing code review
    • fix TODOs/FIXMEs (see sources)
    • send in contributions as patch via mail (address see below)

Contact

  • Vasilios Darlagiannis
  • Roman Schmidt