Analysis of Backoff Protocols for Mulitiple AccessChannels
SIAM Journal on Computing
Charging and accounting for bursty connections
Internet economics
Analysis of practical backoff protocols for contention resolution with multiple servers
Journal of Computer and System Sciences
The Eifel retransmission timer
ACM SIGCOMM Computer Communication Review
Theory, Volume 1, Queueing Systems
Theory, Volume 1, Queueing Systems
Resource pricing and the evolution of congestion control
Automatica (Journal of IFAC)
Optimal sliding-window strategies in networks with long round-trip delays
Computer Networks: The International Journal of Computer and Telecommunications Networking
Preemptive and delay-based mechanisms to provide preference to emergency traffic
Computer Networks and ISDN Systems
Computing with data non-determinism: Wait time management for peer-to-peer systems
Computer Communications
An overhead and resource contention aware analytical model for overloaded Web servers
Journal of Systems and Software
Preemptive and delay-based mechanisms to provide preference to emergency traffic
Computer Networks: The International Journal of Computer and Telecommunications Networking
Cascading multi-way bounded wait timer management for moody and autonomous systems
ICA3PP'11 Proceedings of the 11th international conference on Algorithms and architectures for parallel processing - Volume Part II
ACM SIGMETRICS Performance Evaluation Review
Hi-index | 0.01 |
The notion of timeout (namely, the maximal time to wait before retrying an action) turns up in many networking contexts, such as packet transmission, connection establishment, etc. Usage of timeouts is encountered especially in large-scale networks, where negative acknowledgments (NACKs) on failures have significantly higher delays than positive acknowledgments (ACKs) and frequently are not employed at all. Selection of a proper timeout involves a tradeoff between waiting too long and loading the network needlessly by waiting too little. The common approach is to set the timeout to a large value, such that, unless the action fails, it is acknowledged within the timeout duration with a high probability. This approach is conservative and leads to overly long, far from optimal, timeouts.We take a quantitative approach with the purpose of computing and studying the optimal timeout strategy. The above tradeoff is modeled by introducing a "cost" per unit time (until success) and a "cost" per repeated attempt. The optimal timeout strategy is then defined as one that a selfish user would follow to minimize its expected cost. We discuss the various practical interpretations that these costs may have. We then derive the formulas for the optimal timeout values and study some of their fundamental properties. In particular, we identify the conditions for making parallel attempts from the outset to be worthwhile. In addition, we demonstrate a striking property of positive feedback. This motivates us to study the interaction resulting when many users selfishly apply the optimal timeout strategy; specifically, we use a noncooperative game model and show that it suffers from an inherent instability problem. Some implications of these results on network design are discussed.