Overview

OCEAN (Open Computation Exchange & Auctioning (or Arbitration) Network) is a major ongoing project at the University of Florida's CISE department to develop a fully functional infrastructure supporting the automated, commercial buying and selling of dynamic distributed computing resources over the internet.

The idea is that anyone with spare cycles should be able to deploy an OCEAN server which can run other people's computing tasks for profit, and any developer should be able to easily write a distributed application which any user with a credit card number (or other means of automatic payment) should be able to deploy in distributed fashion using as many suitable OCEAN servers as they can afford to rent for their particular purpose. OCEAN will likely use a distributed, peer-to-peer double-auction mechanism to ensure that jobs are automatically contracted out to the cheapest suitable available bidders, and that OCEAN servers automatically contract themselves out to run the highest-paying available jobs.

 The OCEAN project had its roots at M.I.T. with a group of MIT and former Stanford students, led by Mike Frank. Here is the old OCEAN home page from when it was still based at MIT, and here is an earlier version of the UF page.


 Focus of Research

The main focus of research in OCEAN is:
Investigate self optimizing Peer-to-Peer networks:
OCEAN uses a decentralized peer-to-peer network in order to locate buyers and sellers. Peer-to-peer networks are highly dynamic and change frequently over time. In such a network it is not trivial to locate the best sellers. By optimizing the underlying network structure we would be able to find better matches for buyers and sellers.
Security:
OCEAN will eventually handle real money, hence we must provide a mechanism that will ensure the safety of financial transactions.
Implementation:
We are working on a working prototype to test our results.