Separating Predictable and Unpredictable Flows via Dynamic Flow Mining for Effective Traffic Engineering

Yousuke TAKAHASHI; Keisuke ISHIBASHI; Masayuki TSUJINO; Noriaki KAMIYAMA; Kohei SHIOMOTO; Tatsuya OTOSHI; Yuichi OHSITA; Masayuki MURATA

doi:10.1587/transcom.2017EBT0001

Open Access
Separating Predictable and Unpredictable Flows via Dynamic Flow Mining for Effective Traffic Engineering

Yousuke TAKAHASHI, Keisuke ISHIBASHI, Masayuki TSUJINO, Noriaki KAMIYAMA, Kohei SHIOMOTO, Tatsuya OTOSHI, Yuichi OHSITA, Masayuki MURATA

Full Text Views

100

Cite this

Free PDF (2.2MB)

Summary :

To efficiently use network resources, internet service providers need to conduct traffic engineering that dynamically controls traffic routes to accommodate traffic change with limited network resources. The performance of traffic engineering (TE) depends on the accuracy of traffic prediction. However, the size of traffic change has been drastically increasing in recent years due to the growth in various types of network services, which has made traffic prediction difficult. Our approach to tackle this issue is to separate traffic into predictable and unpredictable parts and to apply different control policies. However, there are two challenges to achieving this: dynamically separating traffic according to predictability and dynamically controlling routes for each separated traffic part. In this paper, we propose a macroflow-based TE scheme that uses different routing policies in accordance with traffic predictability. We also propose a traffic-separation algorithm based on real-time traffic analysis and a framework for controlling separated traffic with software-defined networking technology, particularly OpenFlow. An evaluation of actual traffic measured in an Internet2 network shows that compared with current TE schemes the proposed scheme can reduce the maximum link load by 34% (at the most congested time) and the average link load by an average of 11%.

Publication: IEICE TRANSACTIONS on Communications Vol.E101-B No.2 pp.538-547

Publication Date: 2018/02/01

Publicized: 2017/08/07

Online ISSN: 1745-1345

DOI: 10.1587/transcom.2017EBT0001

Type of Manuscript: PAPER

Category: Internet

Authors

Yousuke TAKAHASHI
  NTT Communications Corporation
Keisuke ISHIBASHI
  NTT Corporation
Masayuki TSUJINO
  NTT Corporation
Noriaki KAMIYAMA
  Fukuoka University
Kohei SHIOMOTO
  Tokyo City University
Tatsuya OTOSHI
  Osaka University
Yuichi OHSITA
  Osaka University
Masayuki MURATA
  Osaka University

Keyword

software defined networking, routing, traffic engineering, macroflow, flow mining

Cite this

Copy

Yousuke TAKAHASHI, Keisuke ISHIBASHI, Masayuki TSUJINO, Noriaki KAMIYAMA, Kohei SHIOMOTO, Tatsuya OTOSHI, Yuichi OHSITA, Masayuki MURATA, "Separating Predictable and Unpredictable Flows via Dynamic Flow Mining for Effective Traffic Engineering" in IEICE TRANSACTIONS on Communications, vol. E101-B, no. 2, pp. 538-547, February 2018, doi: 10.1587/transcom.2017EBT0001.
Abstract: To efficiently use network resources, internet service providers need to conduct traffic engineering that dynamically controls traffic routes to accommodate traffic change with limited network resources. The performance of traffic engineering (TE) depends on the accuracy of traffic prediction. However, the size of traffic change has been drastically increasing in recent years due to the growth in various types of network services, which has made traffic prediction difficult. Our approach to tackle this issue is to separate traffic into predictable and unpredictable parts and to apply different control policies. However, there are two challenges to achieving this: dynamically separating traffic according to predictability and dynamically controlling routes for each separated traffic part. In this paper, we propose a macroflow-based TE scheme that uses different routing policies in accordance with traffic predictability. We also propose a traffic-separation algorithm based on real-time traffic analysis and a framework for controlling separated traffic with software-defined networking technology, particularly OpenFlow. An evaluation of actual traffic measured in an Internet2 network shows that compared with current TE schemes the proposed scheme can reduce the maximum link load by 34% (at the most congested time) and the average link load by an average of 11%.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2017EBT0001/_p

Copy

@ARTICLE{e101-b_2_538,
author={Yousuke TAKAHASHI, Keisuke ISHIBASHI, Masayuki TSUJINO, Noriaki KAMIYAMA, Kohei SHIOMOTO, Tatsuya OTOSHI, Yuichi OHSITA, Masayuki MURATA, },
journal={IEICE TRANSACTIONS on Communications},
title={Separating Predictable and Unpredictable Flows via Dynamic Flow Mining for Effective Traffic Engineering},
year={2018},
volume={E101-B},
number={2},
pages={538-547},
abstract={To efficiently use network resources, internet service providers need to conduct traffic engineering that dynamically controls traffic routes to accommodate traffic change with limited network resources. The performance of traffic engineering (TE) depends on the accuracy of traffic prediction. However, the size of traffic change has been drastically increasing in recent years due to the growth in various types of network services, which has made traffic prediction difficult. Our approach to tackle this issue is to separate traffic into predictable and unpredictable parts and to apply different control policies. However, there are two challenges to achieving this: dynamically separating traffic according to predictability and dynamically controlling routes for each separated traffic part. In this paper, we propose a macroflow-based TE scheme that uses different routing policies in accordance with traffic predictability. We also propose a traffic-separation algorithm based on real-time traffic analysis and a framework for controlling separated traffic with software-defined networking technology, particularly OpenFlow. An evaluation of actual traffic measured in an Internet2 network shows that compared with current TE schemes the proposed scheme can reduce the maximum link load by 34% (at the most congested time) and the average link load by an average of 11%.},
keywords={},
doi={10.1587/transcom.2017EBT0001},
ISSN={1745-1345},
month={February},}

Copy

TY - JOUR
TI - Separating Predictable and Unpredictable Flows via Dynamic Flow Mining for Effective Traffic Engineering
T2 - IEICE TRANSACTIONS on Communications
SP - 538
EP - 547
AU - Yousuke TAKAHASHI
AU - Keisuke ISHIBASHI
AU - Masayuki TSUJINO
AU - Noriaki KAMIYAMA
AU - Kohei SHIOMOTO
AU - Tatsuya OTOSHI
AU - Yuichi OHSITA
AU - Masayuki MURATA
PY - 2018
DO - 10.1587/transcom.2017EBT0001
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E101-B
IS - 2
JA - IEICE TRANSACTIONS on Communications
Y1 - February 2018
AB - To efficiently use network resources, internet service providers need to conduct traffic engineering that dynamically controls traffic routes to accommodate traffic change with limited network resources. The performance of traffic engineering (TE) depends on the accuracy of traffic prediction. However, the size of traffic change has been drastically increasing in recent years due to the growth in various types of network services, which has made traffic prediction difficult. Our approach to tackle this issue is to separate traffic into predictable and unpredictable parts and to apply different control policies. However, there are two challenges to achieving this: dynamically separating traffic according to predictability and dynamically controlling routes for each separated traffic part. In this paper, we propose a macroflow-based TE scheme that uses different routing policies in accordance with traffic predictability. We also propose a traffic-separation algorithm based on real-time traffic analysis and a framework for controlling separated traffic with software-defined networking technology, particularly OpenFlow. An evaluation of actual traffic measured in an Internet2 network shows that compared with current TE schemes the proposed scheme can reduce the maximum link load by 34% (at the most congested time) and the average link load by an average of 11%.
ER -