DistributedResourceAllocation
From P2P Wiki
Definition
One of the main goals of peer-to-peer (P2P) networks is to benefit from sharing resources --data storage, computer cycles or bandwidth-- located at desktop computers. In this context, resource allocation algorithms play an important role in optimizing the use of those massive available assets. These algorithms can be applied to guarantee resource reliability by providing adequate resource redundancy. This reliability may be used for building distributed backup storage systems [1], for distributed computations [2] or in content distribution networks [3] among others.
In this line, incentives mechanisms, trading protocols and trusting algorithms have been used to improve the distributed resource allocation problem. But recently, economic systems are receiving increasing interest in this field. For example, Bittorrent's tit-for-tat [4], PeerMart's[5] virtual currency or the trading mechanisms of several backup storage systems [1],[6] provide interesting solutions to avoid free-riding problems or improve resource reliability in peer-to-peer networks.
The best solutions for resource allocation problems are the algorithms based on reciprocal exchange trades. Inspired in Reciprocal Exchange Networks [7] and barter rings, we believe that these algorithm can provide optimal resource stability while preserving node fairness in a decentralized way. Furthermore, reciprocal exchange can simplify the overall allocation algorithms and thus avoiding the unnecessary complexity of asymmetric trades or currency-based approaches.
References
- M.L. Fakult, "PeerStore: Better Performance by Relaxing in Peer-to-Peer Backup", in proceedings of P2P'04.
- "SETI@home", http://setiathome.berkeley.edu/
- "The Coral Content Distribution Network", http://www.coralcdn.org/
- "BitTorrent", http://www.bittorrent.org/
- David Hausheer, "PeerMart", http://www.peermart.net/
- B.F. Cooper and H. Garcia-Molina, "Peer-to-Peer Data Trading to Preserve Information", in Journal of Information Systems, Volume 20, Num 2, Pages 133-170
- J.P. Stodedr, "Reciprocal Exchange Networks: Implications for Macroeconomic Stability"

