Minimizing Queue Length Regret Under Adversarial Network Models
Name
CV_C_207.pdf
Description
Accepted version
Size
706.23 KB
Format
Adobe PDF
Checksum (MD5)
63e13eb714749a178b7f82a7e3ff5ae9
Author(s)
Liang, Qingkai
Modiano, Eytan H
Date Issued
March 2018
Journal
Proceedings of the ACM on measurement and analysis of computing systems
Publisher
Association for Computing Machinery (ACM)
Citation
Liang, Qingkai and Eytan Modiano. “Minimizing Queue Length Regret Under Adversarial Network Models.” Proceedings of the ACM on measurement and analysis of computing systems, vol. 2, no. 1, 2018, Article 11 © 2018 The Author(s)
Version
Author's final manuscript
Abstract
Stochastic models have been dominant in network optimization theory for over two decades, due totheir analytical tractability. However, these models fail to capture non-stationary or even adversarialnetwork dynamics which are of increasing importance for modeling the behavior of networksunder malicious attacks or characterizing short-term transient behavior. In this paper, we focuson minimizing queue length regret under adversarial network models, which measures the finite-time queue length difference between a causal policy and an “oracle” that knows the future. Twoadversarial network models are developed to characterize the adversary’s behavior. We provide lowerbounds on queue length regret under these adversary models and analyze the performance of twocontrol policies (i.e., the MaxWeight policy and the Tracking Algorithm). We further characterizethe stability region under adversarial network models, and show that both the MaxWeight policyand the Tracking Algorithm are throughput-optimal even in adversarial settings.
MIT Department
Massachusetts Institute of Technology. Laboratory for Information and Decision Systems
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/3179414
