中里 秀則 教授 Hidenori Nakazato, Professor

中里 秀則　教授
NAKAZATO, Hidenori　Professor
学会論文誌
1
Shohei Kamamura, Daisaku Shimazaki, Atsushi Hiramatsu, and Hidenori Nakazato,
“Autonomous IP Fast Rerouting with Compressed Backup Flow Entries using OpenFlow”,
IEICE Trans. INF. ＆ SYST. Vol.E96-D, No. 2, pp. 184-192, February 2013.
【要約】
This paper proposes an IP fast rerouting method which can be implemented in OpenFlow framework. While the current IP is
robust, its reactive and global rerouting processes require the long recovery time against failure. On the other hand, IP fast rerouting provides a milliseconds-order recovery time by proactive and local restoration mechanism. Implementation of IP fast rerouting is not common in real systems, however；it requires the coordination of additional forwarding functions to a commercial
hardware. We propose an IP fast rerouting mechanism using OpenFlow that separates control function from hardware implementation. Our mechanism does not require any extension of current forwarding hardware. On the contrary, increase of backup routes
becomes main overhead of our proposal. We also embed the compression mechanism to our IP fast rerouting mechanism. We
show the effectiveness of our IP fast rerouting in terms of the fast restoration and the backup routes compression effect through
computer simulations.
2
鎌村 星平 , 島崎 大作 , 平松 淳 , 中里 秀則 ,
“動的なパス設定制御に基づく１＋１パスプロテクション維持方式の提案”,
電子情報通信学会論文誌（B）, Vol.J96-B, No. 2, pp. 48-58, 2013年２月 .
【要約】
１＋１パスプロテクション網の保守時間制約を緩和する , 冗長構成を維持する方式を提案する．１＋１パスプロテク
ションでは , 始終点ノード間にノードやリンクを共用しない二つの伝送パスを設定し , 両パスに対してデータを同時
転送することで網の高信頼化を実現する．一方 , 運用保守の観点では , 装置や伝送路の故障により一方のパスが切断
された場合には速やかに故障箇所を復旧することで稼動率目標値を維持する必要があり , 運用保守コストが増加する
課題がある．本論文では , 冗長パス構成が一時的に維持できなくなった際に , 新たな予備パスを動的に設定すること
で可能な限り１＋１パスプロテクション構成を維持するための , 制御アーキテクチャ及び経路計算アルゴリズムを提
案し , 計算機シミュレーションにより有効性を示す．
3
Shohei Kamamura, Daisaku Shimazaki, Atsushi Hiramatsu, and Hidenori Nakazato,
“Loop-free Fast Rerouting considering Double-link Failures”,
IEICE Trans. Commun., Vol.E95-B, No.12,pp. 3811-3821, December 2012.
【要約】
IP fast rerouting has widely been studied for realizing millisecond-order recovery on pure IP networks. This paper proposes IP
fast rerouting using backup topologies against concurrent double failures. The main issue in recovering from multiple failures is
avoiding forwarding loops. To avoid forwarding loops, we propose a deterministic forwarding algorithm, which estimates the
concurrently occurring failures from the packet header information. We also propose an efficient backup topology design algorithm which is both loop-free and which reduces the number of backup topologies. Our key idea is preparing the adequate diversity of backup routes for arbitrary source and destination pairs by combination of backup topologies. For efficient computation
of diverse routes, we propose a similarity comparison-based algorithm between the original topology and the backup topologies.
Our algorithm can achieve nearly optimal loop-free restoration from double failures on realistic topologies without explicit failure
notification.
85
国際会議
4
Dinh Nguyen and Hidenori Nakazato,
“Rarest-first and Coding Are Not Enough”,
Proc. of Globecom 2012, pp.2707-2712, Anaheim, CA, USA, December 2012.
【要約】
Network coding has been applied successfully in peer- to-peer（P2P）systems to shorten the distribution time. Pieces of data, i.e.
blocks, are combined, i.e. encoded, by the sending peers before forwarding to other peers. Even though requiring all peers to
encode might achieve shortest distribution time, it is not necessarily optimal in terms of computational resource consumption.
Short finish time, in many cases, can be achieved with just a subset of carefully chosen peers. P2P systems, in addition, tend to
be heterogeneous in which some peers, such as hand-held devices, would not have the required capacity to encode. We therefore
envision a P2P system where some peers encode to improve distribution time and other peers, due to limited computational capacity or due to some system-wide optimization, do not encode. Such a system gives rise to a block- selection problem which has
never happened in both pure non- coding and full network coding-enabled P2P systems. We identify the problem and fix the current block-selection algorithm to address it. Simulation evaluation confirms the effectiveness of our proposed algorithm without
which the system performance degrades considerably.
5
Huan Wang and Hidenori Nakazato,
“Failure detection in P2P-Grid environments”,
In 32nd International Conference on Distributed Computing Systems Workshops, June 2012.
【要約】
P2P-Grid system provides a framework for converging Grid and peer-to-peer network to deploy large-scale distributed applications. However, nodes dynamic participation arbitrarily makes failure more common than in other systems. As the most common
technique for fault tolerance, Checkpointing-and-Recovery saves application execution state during normal execution and restoring the saved state after a failure to reduce the amount of lost work. In this paper, we propose a Checkpointing-and-Recovery
architecture for applications restarting as soon as possible on P2P-Grid systems. And a failure detection mechanism is a necessary
prerequisite to fault tolerance and fault recovery in P2P-Grid system. To this end, failure detection mechanisms as an integral
part of P2P-Grid system have been well-studied. We investigate how the design of various failure detection algorithms affects
their performance in node average failure detection time. We also provide numerical results based on both theoretical analysis and
simulations. The evaluated results show improvement of the performance on the basis of the WP failure detection algorithm.
国内研究会
6
Dinh Nguyen and Hidenori Nakazato,
“Network Coder Placement for Peer-to-Peer Content Distribution”,
Technical Report of IEICE, vol. 112, no. 309, CS2012-74, pp.59-64 , November 2012.
【要約】
Network coding has been shown to accelerate peer-to-peer content distribution. Coding, i.e. combining data, at intermediate
peers, can eliminate content duplication transmitted to downstream peers, which results in much shorter distribution time. The
question is: do we need to code at every peer to achieve such good performance. In this paper, we figure the condition for network
coding to be effective, and based on which, we propose an algorithm to selectively place coders inside a peer-to-peer content distribution in order to shorten distribution time, yet, using far fewer peers as encoders. Simulation results confirm the efficiency of
our placement method.
7
菊地 俊介 , 中里 秀則 ,
“コンテンツの長さに注目したネットワークキャッシュ制御手法の提案と評価”,
信学技報 , vol.112, no. 309, CS2012-73, pp.53-58, 2012年11月 .
【要約】
新世代ネットワークのネットワーク仮想化によりアプリケーションに特化したネットワークを構築可能になる．そこ
でキャッシュもアプリケーションに合わせた特性チューニングが可能であれば , アプリケーションによっては現状よ
りレスポンスタイムを改善できる可能性がある．本稿では , コンテンツの長さに注目し , アプリケーションに適した
コンテンツ長でキャッシュを制御することで , レスポンスタイムの向上といった実質的なユーザの使い勝手を改善す
86
ることができるネットワークキャッシュ制御方式について提案する．また数値計算モデルとシミュレーションにより
方式を評価し , 従来のキャッシュ方式に対して遅延特性が改善していることを確認した．
著書
8
Hidehiro Kanemitsu, Gilhyon Lee, Hidenori Nakazato, Takashige Hoshiai, and Yoshiyori Urano,
“On the effect of applying The task clustering for identical processor utilization to heterogeneous
systems”,
In Soha Maad, editor, “Grid Computing - Technology and Applications, Widespread Coverage and New Horizons”,
chapter 2, pages 29-46. InTech, 2012.
社会活動・学会活動
9
電子情報通信学会東京支部委員（2011-2013）
10
IEEE Region 10 Secretary（2013-2014）
11
IEEE MGA IT Coordination and Oversight Committee（2012-2013）
12
IEEE Waseda Student Branch Counselor（2010-）
13
IEEE GLOBECOM プログラム委員（2008-2013）
14
IEEE ICC プログラム委員（2005-2013）
15
The Twelfth International Workshop on Assurance in Distributed Systems and Networks（ADSN 2013）プログラム副
16
The Eleventh International Symposium on Autonomous Decentralized System（ISADS 2013）プログラム委員
17
IEEE International Communications Quality and Reliability（CQR）Workshop プログラム委員（2007-2013）
委員長
朴 容震　教授（任期付）
PARK, Yong-Jin　Professor
国際会議
1
Byeong-Gi Kim, Mu-Roung Chen, Sho Harada, Yuki Kojima, Yong-Jin Park,
“Content-Centric Information Collection for Cyber-Physical Systems”,
AsiaFI 2012 Summer School in Kyoto University, Kyoto, Japan, 2012. 8. 22.
【要約】
Content-Centric Networking（CCN）gives new network architecture that enables routing by named content rather than named
host. On the other hand, Cyber Physical Systems（CPS）has been introduced as the monitored, coordinated, controlled, and
integrated systems between cyber and physical elements. In CPS, many sensing information are collected. CCN is well-suited to the
needs of such systems to support the information collections. In this letter, we propose an effective Content-Centric Information
Collection for Cyber Physical Systems（CCIC）by extending CCN architecture. The proposal aims to apply CCN to CPS
applications with flexible data requests and efficient use of network resources.
87