In recent years, the time variation of Internet traffic has increased due to the growth of streaming and cloud services. Backbone networks must accommodate such traffic without congestion. Traffic engineering with traffic prediction is one approach to stably accommodating time-varying traffic. In this approach, routes are calculated from predicted traffic to avoid congestion, but predictions may include errors that cause congestion. We propose prediction-based traffic engineering that is robust against prediction errors. To achieve robust control, our method uses model predictive control, a process control method based on prediction of system dynamics. Routes are calculated so that future congestion is avoided without sudden route changes. We apply calculated routes for the next time slot, and observe traffic. Using the newly observed traffic, we again predict traffic and re-calculate the routes. Repeating these steps mitigates the impact of prediction errors, because traffic predictions are corrected in each time slot. Through simulations using backbone network traffic traces, we demonstrate that our method can avoid the congestion that the other methods cannot.
Tatsuya OTOSHI
Osaka University
Yuichi OHSITA
Osaka University
Masayuki MURATA
Osaka University
Yousuke TAKAHASHI
NTT Corporation
Noriaki KAMIYAMA
NTT Corporation
Keisuke ISHIBASHI
NTT Corporation
Kohei SHIOMOTO
NTT Corporation
Tomoaki HASHIMOTO
Osaka University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Tatsuya OTOSHI, Yuichi OHSITA, Masayuki MURATA, Yousuke TAKAHASHI, Noriaki KAMIYAMA, Keisuke ISHIBASHI, Kohei SHIOMOTO, Tomoaki HASHIMOTO, "Traffic Engineering Based on Model Predictive Control" in IEICE TRANSACTIONS on Communications,
vol. E98-B, no. 6, pp. 996-1007, June 2015, doi: 10.1587/transcom.E98.B.996.
Abstract: In recent years, the time variation of Internet traffic has increased due to the growth of streaming and cloud services. Backbone networks must accommodate such traffic without congestion. Traffic engineering with traffic prediction is one approach to stably accommodating time-varying traffic. In this approach, routes are calculated from predicted traffic to avoid congestion, but predictions may include errors that cause congestion. We propose prediction-based traffic engineering that is robust against prediction errors. To achieve robust control, our method uses model predictive control, a process control method based on prediction of system dynamics. Routes are calculated so that future congestion is avoided without sudden route changes. We apply calculated routes for the next time slot, and observe traffic. Using the newly observed traffic, we again predict traffic and re-calculate the routes. Repeating these steps mitigates the impact of prediction errors, because traffic predictions are corrected in each time slot. Through simulations using backbone network traffic traces, we demonstrate that our method can avoid the congestion that the other methods cannot.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E98.B.996/_p
Copy
@ARTICLE{e98-b_6_996,
author={Tatsuya OTOSHI, Yuichi OHSITA, Masayuki MURATA, Yousuke TAKAHASHI, Noriaki KAMIYAMA, Keisuke ISHIBASHI, Kohei SHIOMOTO, Tomoaki HASHIMOTO, },
journal={IEICE TRANSACTIONS on Communications},
title={Traffic Engineering Based on Model Predictive Control},
year={2015},
volume={E98-B},
number={6},
pages={996-1007},
abstract={In recent years, the time variation of Internet traffic has increased due to the growth of streaming and cloud services. Backbone networks must accommodate such traffic without congestion. Traffic engineering with traffic prediction is one approach to stably accommodating time-varying traffic. In this approach, routes are calculated from predicted traffic to avoid congestion, but predictions may include errors that cause congestion. We propose prediction-based traffic engineering that is robust against prediction errors. To achieve robust control, our method uses model predictive control, a process control method based on prediction of system dynamics. Routes are calculated so that future congestion is avoided without sudden route changes. We apply calculated routes for the next time slot, and observe traffic. Using the newly observed traffic, we again predict traffic and re-calculate the routes. Repeating these steps mitigates the impact of prediction errors, because traffic predictions are corrected in each time slot. Through simulations using backbone network traffic traces, we demonstrate that our method can avoid the congestion that the other methods cannot.},
keywords={},
doi={10.1587/transcom.E98.B.996},
ISSN={1745-1345},
month={June},}
Copy
TY - JOUR
TI - Traffic Engineering Based on Model Predictive Control
T2 - IEICE TRANSACTIONS on Communications
SP - 996
EP - 1007
AU - Tatsuya OTOSHI
AU - Yuichi OHSITA
AU - Masayuki MURATA
AU - Yousuke TAKAHASHI
AU - Noriaki KAMIYAMA
AU - Keisuke ISHIBASHI
AU - Kohei SHIOMOTO
AU - Tomoaki HASHIMOTO
PY - 2015
DO - 10.1587/transcom.E98.B.996
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E98-B
IS - 6
JA - IEICE TRANSACTIONS on Communications
Y1 - June 2015
AB - In recent years, the time variation of Internet traffic has increased due to the growth of streaming and cloud services. Backbone networks must accommodate such traffic without congestion. Traffic engineering with traffic prediction is one approach to stably accommodating time-varying traffic. In this approach, routes are calculated from predicted traffic to avoid congestion, but predictions may include errors that cause congestion. We propose prediction-based traffic engineering that is robust against prediction errors. To achieve robust control, our method uses model predictive control, a process control method based on prediction of system dynamics. Routes are calculated so that future congestion is avoided without sudden route changes. We apply calculated routes for the next time slot, and observe traffic. Using the newly observed traffic, we again predict traffic and re-calculate the routes. Repeating these steps mitigates the impact of prediction errors, because traffic predictions are corrected in each time slot. Through simulations using backbone network traffic traces, we demonstrate that our method can avoid the congestion that the other methods cannot.
ER -