Project ID:
RFP-2007-006
Title:
Classification of Distributed Hash Table methods,
and their suitability to various application domains
Summary:
Various Distributed Hash Table (DHT) methods exist, including Bamboo,
Chord, Pastry, Tapestry, and others. They have been applied to a variety
of domains, including routing, storage, multicast, tunneling, and others.
However, we lack a comprehensive classification method to determine
which methods are most appropriate for the various application domains.
The question is, can a classification method be developed, that
characterizes the various DHT techniques, and determines for which
uses they are most suited?
Full Description:
We seek to advance the understanding of existing DHT techniques,
and their applicability to various application domains.
This might include characterization and comparison of existing
methods (e.g., Bamboo, Chord, Pastry, Tapestry, etc.) on their performance,
scaling, and security properties.
Are there existing, or new, techniques to classify DHT methods, which
help understand where the methods are most useful? Do fundamental
issues exist when trying to apply different methods to different application
areas (e.g. routing, storage, multicast, tunneling, etc.)?
If fundamental classification methods are discovered, can the methods
help predict other classes of DHTs, and where they might be used?
What are directions for, and pace of, evolving work in this area?
Since Cisco engineers have widely varying exposure to DHTs,
it might also be helpful to highlight the most important work in the
area, and assemble a list like, "The (10?) most important things
Cisco engineers should know about DHTs".
Areas of interest might include:
How many neighbors are optimal?
How often do updates occur?
What level of reliability is achievable?
How often should the finger table be checked to see if the peers are alive?
What percentage of the nodes need to be compromised to defeat the
security of the system?
Constraints and other information:
IPR will stay with the University. Cisco expects customary scholarly dissemination of results, and hopes that promising results would be made available to the community without limiting licenses, royalties, or other encumbrances.
Proposal submission:
Cisco is not currently accepting proposals for this RFP.
Questions? Contact: research@cisco.com
