Optimization flow control—I: basic algorithm and convergence
IEEE/ACM Transactions on Networking (TON)
An Incentive Mechanism for P2P Networks
ICDCS '04 Proceedings of the 24th International Conference on Distributed Computing Systems (ICDCS'04)
Limited reputation sharing in P2P systems
EC '04 Proceedings of the 5th ACM conference on Electronic commerce
ECHOS: edge capacity hosting overlays of nano data centers
ACM SIGCOMM Computer Communication Review
Distributed resource allocation algorithms for peer-to-peer networks
Proceedings of the 3rd International Conference on Performance Evaluation Methodologies and Tools
Uplink allocation beyond choke/unchoke: or how to divide and conquer best
CoNEXT '08 Proceedings of the 2008 ACM CoNEXT Conference
TCPeer: rate control in P2P over IP networks
ITC20'07 Proceedings of the 20th international teletraffic conference on Managing traffic performance in converged networks
Incentives for large peer-to-peer systems
IEEE Journal on Selected Areas in Communications
Mathematical decomposition techniques for distributed cross-layer optimization of data networks
IEEE Journal on Selected Areas in Communications
Hi-index | 0.00 |
In peer-to-peer networks, each peer plays the role of client and server. As a server, it receives content requests of others and decides to what extent it will satisfy them by allocating upload bandwidth. As a client, it sends its own requests to others to download content. We consider a star topology network with the capacity bottleneck being the peer access link to the backbone. Peers are assumed to be cooperative but with different resource needs and valuations, captured by their utility function, as well as with different service capabilities. We consider the problem of maximizing total network utility (social welfare) through controlling the bandwidth portion allocated to downloaders and uploaders of each peer. For autonomous unsupervised network operation, the fundamental challenge is to achieve the goal above in a distributed fashion. However, the global bandwidth sharing problem is difficult to solve in a decentralized fashion because of coupled utility functions and constraints, since the utility of a peer depends on allocation decisions of others. By defining new auxiliary variables and constraints, we transform the problem into one that is amenable to distributed solution by dual decomposition methods. The iterative algorithm involves solving separate optimization problems by each peer and updating Lagrange multipliers. Interestingly, the Lagrangian multipliers corresponding to new constraints can be interpreted as metrics for the level of pairwise resource contribution. The solution of the social welfare problem consists of the optimal pairwise amounts of allocated bandwidth and the optimal dual variables that induce these allocations. This leads to a meaningful lightweight algorithm which converges to the globally optimum bandwidth allocation.