Closure and Convergence: A Foundation of Fault-Tolerant Computing
IEEE Transactions on Software Engineering - Special issue on software reliability
Providing guaranteed services without per flow management
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
Self-stabilizing systems in spite of distributed control
Communications of the ACM
A General Theory of Constrained Max-Min Rate Allocation for Multicast Networks
ICON '00 Proceedings of the 8th IEEE International Conference on Networks
Minimum Rate Guarantee without Per-Flow Information
ICNP '99 Proceedings of the Seventh Annual International Conference on Network Protocols
GTP: group transport protocol for lambda-Grids
CCGRID '04 Proceedings of the 2004 IEEE International Symposium on Cluster Computing and the Grid
A queueing analysis of max-min fairness, proportional fairness and balanced fairness
Queueing Systems: Theory and Applications
Max-Min-Fair Best Effort Flow Control in Network-on-Chip Architectures
ICCS '08 Proceedings of the 8th international conference on Computational Science, Part I
Scalable quality of service across multiple domains
Computer Communications
Scalable max-min fairness in wireless ad hoc networks
Ad Hoc Networks
Convergence rate control for distributed multi-hop wireless mesh networks
Computers and Electrical Engineering
Hi-index | 5.23 |
Let a flow be a sequence of packets that are sent from a source computer to a destination computer. In this paper, we consider the fair allocation of bandwidth to each flow in a computer network. We focus on max-min fairness, which assigns to each flow the largest possible bandwidth that avoids affecting other flows. What distinguishes our approach is that routers only maintain a constant amount of state, i.e., no per-flow state is maintained. This is consistent with trends in the Internet (such as the proposed Differentiated Services Internet architecture). In addition, to provide a high degree of fault-tolerance, we ensure our approach is self-stabilizing, that is, it returns to a normal operating state after a finite sequence of faults.