CaltechAUTHORS
  A Caltech Library Service

A Control-Theoretic Approach to In-Network Congestion Management

Wu, Ning and Bi, Yingjie and Michael, Nithin and Tang, Ao and Doyle, John C. and Matni, Nikolai (2018) A Control-Theoretic Approach to In-Network Congestion Management. IEEE/ACM Transactions on Networking, 26 (6). pp. 2443-2456. ISSN 1063-6692. http://resolver.caltech.edu/CaltechAUTHORS:20181005-093939759

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20181005-093939759

Abstract

WANs are often over-provisioned to accommodate worst-case operating conditions, with many links typically running at only around 30% capacity. In this paper, we show that in-network congestion management can play an important role in increasing network utilization. To mitigate the effects of in-network congestion caused by rapid variations in traffic demand, we propose using high-frequency traffic control (HFTraC) algorithms that exchange real-time flow rate and buffer occupancy information between routers to dynamically coordinate their link-service rates. We show that the design of such dynamic link-service rate policies can be cast as a distributed optimal control problem that allows us to systematically explore an enlarged design space of in-network congestion management algorithms. This also provides a means of quantitatively comparing different controller architectures: we show, perhaps surprisingly, that centralized control is not always better. We implement and evaluate HFTraC in the face of rapidly varying UDP and TCP flows and in combination with AQM algorithms. Using a custom experimental testbed, a Mininet emulator, and a production WAN, we show that HFTraC leads to up to 66% decreases in packet loss rates at high link utilizations as compared to FIFO policies.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/tnet.2018.2866785DOIArticle
ORCID:
AuthorORCID
Wu, Ning0000-0002-2943-7711
Tang, Ao0000-0001-6296-644X
Doyle, John C.0000-0002-1828-2486
Matni, Nikolai0000-0003-4936-3921
Additional Information:© 2018 IEEE. Manuscript received September 22, 2017; revised May 11, 2018; accepted July 23, 2018; approved by IEEE/ACM TRANSACTIONS ON NETWORKING. Editor C. Joo. Date of publication October 2, 2018; date of current version December 14, 2018. This work was supported in part by NSF, in part by AFOSR, in part by Google, and in part by Huawei.
Funders:
Funding AgencyGrant Number
NSFUNSPECIFIED
Air Force Office of Scientific Research (AFOSR)UNSPECIFIED
GoogleUNSPECIFIED
HuaweiUNSPECIFIED
Subject Keywords:Network congestion management, distributed control, optimal control
Record Number:CaltechAUTHORS:20181005-093939759
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20181005-093939759
Official Citation:N. Wu, Y. Bi, N. Michael, A. Tang, J. C. Doyle and N. Matni, "A Control-Theoretic Approach to In-Network Congestion Management," in IEEE/ACM Transactions on Networking, vol. 26, no. 6, pp. 2443-2456, Dec. 2018. doi: 10.1109/TNET.2018.2866785
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:90141
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:05 Oct 2018 23:22
Last Modified:20 Dec 2018 17:12

Repository Staff Only: item control page