Advance

KASHIWAZAKI Hiroki

    Department of Informatics　Associate Professor
Last Updated :2024/04/25

Researcher Information

Degree

  • Master of Engineering（Hokkaido University）
  • Doctor of Information Science（Hokkaido University）

URL

Research funding number

  • 80422004

J-Global ID

Profile

  • Assistant Professor, Tokyo University of the Arts. Slashdot Japan Editor. Interests: autonomous distribution, artificial life, animation. Love: Ruby, BSD.

Research Interests

  • moe   cloud computing   traffic engineering   人工生命   ネットワークルーティング   Artificial Life   Network Routing   

Research Areas

  • Humanities & social sciences / Aesthetic practices
  • Humanities & social sciences / Aesthetics and art studies
  • Informatics / Information networks
  • Informatics / Computer systems
  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering) / Communication and network engineering

Academic & Professional Experience

  • 2022/09 - Today  Kindai UniversityFaculty of Informatics Department of Informatics准教授
  • 2019/04 - 2022/09  National Institute of InformaticsProject Associate Professor
  • 2019/04 - Today  Osaka universityCybermedia CenterGuest Associate Professor
  • 2016/06 - 2019/03  Osaka UniversityCybermedia CenterAssociate Professor
  • 2012/12 - 2016/06  Osaka UniversityCybermedia CenterAssistant Professor
  • 2010/01 - 2012/11  Tokyo University of the ArtsAssistant Professor
  • 2005/06 - 2009/12  Hokkaido UniversityAssistant Professor

Education

  • 2001/04 - 2005/05  Hokkaido University  Graduate School of Engineering  電子情報工学専攻
  • 1995/04 - 1999/03  Hokkaido University  Faculty of Engineering

Association Memberships

  • ACM   地域間相互接続プロジェクト   人工知能学会   電子情報通信学会   IEEE   情報処理学会   Association for Computing Machinery   Regional Internet BackBone   The Japanese Society for Artificial Intelligence   Information and Communication Engineers   The Institute of Electronics   Inc.   The Institute of Electrical and Electronics Engineers   Information Processing Society of Japan   

Published Papers

  • Shuichiro Shimatani; Hiroki Kashiwazaki; Iguchi Nobukazu
    2023 IEEE 47th Annual Computers, Software, and Applications Conference (COMPSAC) IEEE 518 - 525 2023/06 [Refereed]
  • Daisuke Sasaki; Hiroki Kashiwazaki; Mitsuhiro Osaki; Kazuma Nishiuchi; Ikuo Nakagawa; Shunsuke Kikuchi; Yutaka Kikuchi; Shintaro Hosoai; Hideki Takase
    2023 IEEE 47th Annual Computers, Software, and Applications Conference (COMPSAC) IEEE 1133 - 1140 2023/06 [Refereed]
  • Kazuhiro Mishima; Hiroki Kashiwazaki; Yoshiaki Kitaguchi
    Proceedings of the 2023 ACM SIGUCCS Annual Conference ACM 42 - 47 2023/03 [Refereed]
  • Hiroki Kashiwazaki; Hideo Masuda; Kazuhiro Mishima; Kensuke Miyashita
    Proceedings of the 2022 ACM SIGUCCS Annual Conference ACM 3 - 6 2022/03 [Refereed]
  • Hiroki Kashiwazaki; Takuro Ozaki; Hajime Shimada; Yusuke Komiya; Eisaku Sakane; Kazuhiro Mishima; Shiu Sakashita; Nariyoshi Yamai; Yoshiaki Kitaguchi; Kensuke Miyashita
    ACM SIGUCCS Annual Conference ACM 54 - 59 2021/03 [Refereed]
  • Hiroki Kashiwazaki; Hiroki Takakura; Shinji Shimojo
    2021 International Conference on Information Networking (ICOIN) IEEE 143 - 148 2021/01 [Refereed]
  • A proposal of autonomous wireless mesh network with autonomous mobile robot for sustainable smart-rural
    Hiroki Kashiwazaki
    研究報告インターネットと運用技術(IOT) 2020-IOT-51 (2) 1 - 4 2020/09
  • Hiroki Kashiwazaki; Hiroki Takakura; Shinji Shimojo
    2020 IEEE 44th Annual Computers, Software, and Applications Conference (COMPSAC) IEEE 607 - 612 2020/07 [Refereed]
  • Qualitative and quantitative performance evaluations of relatively inexpensive storage products (2)
    Hiroki Kashiwazaki
    研究報告インターネットと運用技術(IOT) 2020-IOT-50 (7) 1 - 6 2020/06
  • COVID-19が蔓延してみたらCIOより上手にDXが進化した件
    尾崎 拓郎; 嶋田 創; 柏崎 礼生; 岡部 寿男; 北口 善明
    研究報告インターネットと運用技術(IOT) 2020-IOT-49 (4) 1 - 1 2020/05
  • 柏崎礼生; 坂根栄作; 込山悠介; 宮崎純; 中沢実; 岡部寿男
    情報処理学会研究報告(Web) 2020-IOT-49 (7) 1 - 8 2020/05
  • 嶋田創; 柏崎礼生; 池部実; 北口善明; 山井成良; 坂下秀; 村上登志男; 石島悌; 石橋勇人; 佐藤聡; 中村豊; 林治尚; 松本直人; 三島和宏; 大谷誠; 桝田秀夫; 宮下健輔
    情報処理学会研究報告(Web) 2020-IOT-49 (5) 1 - 8 2020/05
  • 山岳地域における持続可能性のあるインターネット接続性の試行
    柏崎 礼生
    ITRC technical report 1343-3083 2020/04
  • A proposal of quantitative approaches to evaluate a resilience of wide-area distributed sysmtems. abstract
    Kashiwazaki Hiroki
    ITRC technical report 1343-3083 2020/04
  • 【最新】2019春夏ニーソックストレンド総まとめ
    柏崎 礼生
    ITRC technical report 1343-3083 2020/04
  • Distcloud v3の設計と実装
    Kashiwazaki Hiroki
    ITRC technical report 1343-3083 2020/04
  • Design, implementation and its evaluation of a multi-agent simulator using Golang
    Hiroki Kashiwazaki
    研究報告インターネットと運用技術(IOT) 2020-IOT-48 (5) 1 - 3 2020/03
  • Hiroki Kashiwazaki; Hiroki Takakura; Shinji Shimojo
    2019 International Conference on Information and Communication Technologies for Disaster Management (ICT-DM) IEEE 1 - 8 2019/12
  • Hiroki Kashiwazaki; Hiroki Takakura; Shinji Shimojo
    インターネットと運用技術シンポジウム論文集 2019 17 - 24 2019/12 [Refereed]
     
    A wide-area distributed application is affected by network failure due to natural disasters because the servers on which the application operates are distributed geographically in a wide area. Failure Injection Testing (FIT) is a method for verifying fault tolerance of widely distributed applications. In this paper, by limiting network failures only to the connection line, whole FIT scenarios are generated, and exhaustive evaluation of fault tolerance is performed. The authors propose a method to omit the evaluations from the aspect of topological constraint conditions. And they evaluate the visualization method of performance data obtained from this evaluation and the reduction of the fault tolerance evaluation cost by the proposed method.
  • Yuki Fujita; Atsuo Inomata; Hiroki Kashiwazaki
    2019 20th Asia-Pacific Network Operations and Management Symposium (APNOMS) IEEE 1 - 4 2019/09 [Refereed]
  • Hiroki Kashiwazaki; Shinnosuke Miura; Shinji Shimojo
    2019 IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC) IEEE 36 - 41 2019/07 [Refereed]
     
    A wide area distributed application is affected by network failure due to natural disasters because the servers on which the application operates are distributed geographically in a wide area. Failure Injection Testing (FIT) is a method for verifying fault tolerance of widely distributed applications. In this paper, by limiting network failures to the connection line, whole FIT scenarios are generated and exhaustive evaluation of fault tolerance is performed. Authors evaluate the visualization method of performance data obtained from this evaluation and the reduction of the fault tolerance evaluation cost by the proposed method.
  • Evaluation of wide-area distributed services by SDN-FIT system
    Hiroki Kashiwazaki; Shinnosuke Miura; Hiroki Takakura; Shinji Shimojo
    Proceedings of the 48th meeting of the Asia-Pacific Advanced Network 2 - 11 2227-3026 2019/07 [Refereed]
  • Yuki Fujita; Atsuo Inomata; Hiroki Kashiwazaki
    2019 20th Asia-Pacific Network Operations and Management Symposium (APNOMS) IEEE 2019-IOT-46 (7) 1 - 5 2188-8787 2019/06 
    As the number of Internet users increases, their usage also diversifies, and it is important to prevent Identity on the Internet (Digital Identity) from being violated. Unauthorized authentication is one of the methods to infringe Digital Identity. Multi-factor authentication has been proposed as a method for preventing unauthorized authentication. However, the cryptographic authenticator required for multi-factor authentication is expensive both financially and UX-wise for the user. In this paper, we design, implement and evaluate multi-factor authentication using MyNumberCard provided by public personal identification service and WebUSB, which is being standardized.
  • Shinnosuke Miura; Hiroki Kashiwazaki
    研究報告インターネットと運用技術（IOT） 2019-IOT-45 (15) 1 - 8 2188-8787 2019/05
  • A certain personal information leak in a university, and its cleanup
    Hiroki Kashiwazaki
    SIGUCCS '18 Proceedings of the 46th annual ACM SIGUCCS fall conference 2018/11 [Refereed]
  • 柏崎 礼生
    SIGUCCS '18 Proceedings of the 2018 ACM on SIGUCCS Annual Conference 43 - 50 2018/10 [Refereed]
     
    In December 2017, a Japanese university announced a large-scale personal information leak. According to public sources, the leak was caused on several systems by several points of unauthorized access on several systems. Additionally in February 2018, a Japanese research institute announced a large-scale security incident. In Japan, many reports of cyber security incidents are announced for a year. After security incidents occur, a supervisory agency (in these cases, the Ministry of Education, Culture, Sports, Science and Technology a.k.a MEXT) and security consulting firms order the institutes to "strengthen governance in their own institutes'', "develop preventive measures'' and even "buy (expensive) security appliances''. The purpose of this paper is to share the example of a Japanese university's information leak and its cleanup process, as well as to explore what "governance'' means in higher education and research institutes, and how we operate Computer Emergency Response Teams / Computer Security Incident Response Teams.
  • Hiroki Kashiwazaki
    研究報告インターネットと運用技術（IOT） 2018-IOT-42 (7) 1 - 4 2018/06 
    An author is a spider-like creature who has decided that, since he is intelligent, he will resist the instincts which lead his species through an exceptionally violent life-cycle which they call “the Plan” - including a drop in intelligence during “winter”. Last December, a certain university announced large-scale personal information leak caused on several systems by several unauthorized accesses. Security advisory consulting companies order the university to “strengthen governance in their own institutes”. This paper shows an aftermath of the personal information leak incident in the university and how administration bureau deals with a lot of the vulnerability in the university.
  • Hiroki Kashiwazaki
    研究報告インターネットと運用技術（IOT） 2018-IOT-41 (25) 1 - 4 2018/05 
    In December 2017, a Japanese university announced large-scale personal information leak. According to open information, the leak was caused on several systems by several unauthorized accesses. Also in February 2018, a Japanese research institute announced a large-scale security incident. Even only in Japan, many reports of cyber security incidents are announced for a year. After security incidents occur, a supervisory agency (in these cases, Ministry of Education, Culture, Sports, Science and Technology a.k.a MEXT) and security advisory consulting companies order the institutes to “strengthen governance in their own institutes”. This paper shows an example of personal information leak incident in an university and a beast that asked what is the governance in the university at the heart of the university.
  • Dissipative Infrastructure, Design and Implementation
    柏崎 礼生
    ITRC Technical Report 2018/05
  • RICC update: RICC における減災への取り組み
    柏崎 礼生
    2018/05
  • ``Chaos Infrastructure'' for trustworthy transactions of data harvesting
    Hiroki Kashiwazaki
    ITRC Technical Report 2018/05
  • Hiroki Kashiwazaki
    2018-IOT-40 (4) 1 - 4 2018/03 
    The researchers engaging to wide-area geographically distributed platform “Distcloud” have kept maintaining and operating the platform since the start of Distcloud, 2012. They have already lost their enthsiasm for expasion of Distcloud and development of Distcloud and DESTCloud that evaluate and validate fault tolerance of Distcloud have been slugging for a long time. Although an author have also sit around for years, he started to exert himself to the universe and his childfood dream again.
  • Hiroki Kashiwazaki
    研究報告インターネットと運用技術（IOT） 2017-IOT-38 (15) 1 - 6 2017/06 
    ストレージの性能評価には様々な手法がある．本稿ではテラバイト単価 100 USD 程度の比較的安価なストレージ製品を定性的，定量的に評価する．本稿は費用対効果を示すとともに，持続可能でベンダー横断的な評価へと展開する展望について述べる．
  • Hiroki Kashiwazaki
    研究報告インターネットと運用技術（IOT） 2017-IOT-37 (13) 1 - 6 2017/05 
    People can obtain a huge and a strong power of computing resources by a raise of cloud computing environments. Pay as you go is one of the identical definition of cloud computing environments. If a person uses the large amount of cloud computing resources, then he must be charged. Meanwhile a person sometimes can meet the great amount of bill and then be embarrassed. This paper shows a pitiful person who were charged more than 50 thousands dollars as one month usage fee by a certain cloud computing provider. And then also the paper explains the reason of the incident, and proposes the model that can be solve this sort of contradiction between users and providers.
  • 山井成良; 坂下秀; 吉田健一; 石橋圭介; 宮下健輔; 柏崎礼生
    研究報告インターネットと運用技術（IOT） 2017-IOT-36 (40) 1 - 1 2188-8787 2017/03 
    In this special session, the chiefs and leaders of IPSJ SIG-IOT and IEICE IA discuss the future of special interest groups and academic societies.
  • 柏崎 礼生; 藤本 祥人; 宮永 勢次; 森原 一郎
    インターネットと運用技術シンポジウム2016論文集 2016 (2016) 3 - 10 2016/12 [Refereed]
     
    Sometime it may be necessary for a virtualization infrastructure that is designed to server virtualization and consolidation to reinforcement its computing resources triggered by a growth of its requirement and limitation of its support time. This paper shows parameters that are required to design quantitative reinforcement and evaluate its cost of implementation and operation referring to an example of a virtualization infrastructure supporting about 100 VMs.
  • 柏崎礼生; 西内一馬; 北口善明; 市川昊平; 近堂徹; 中川郁夫; 菊池豊
    インターネットと運用技術シンポジウム2016論文集 2016 (2016) 33 - 40 2016/12 [Refereed]
     
    To improve a business continuity plan to more realistic one, it is necessary for the business that include processes of information communication technology system to conduct trainings periodically. Trainings needs their virtual scenarios that emulate evaluate various faults. The costs to generate scenarios can not be ignored in large scale, wide area distributed system that consists of many, various organizations. This paper shows designs and implementations of interfaces to reduce generating costs for network disaster trainings. Evaluations of effectiveness of the interface are also discussed.
  • 柏崎礼生; 北口善明; 市川昊平; 近堂徹; 中川郁夫; 菊池豊; 下條真司
    インターネットと運用技術シンポジウム2016論文集 2016 (2016) 18 - 25 2016/12 [Refereed]
     
    There are two approach to demonstrate experiments of wide area distributed appliactions. One is to demonstrate in virtual environments that emulate wide area distributed environments. The other is to demonstrate in real environments that consists of geographically wide are distributed computing resources and communication lines. Although the real environment can be persuasive to evaluate the experiments, the cost to install the environment can not be negligible. This paper shows a design and operational costs of a real wide area distributed environments with 11 domestic regions and one international region.
  • 坂下秀; 柏崎礼生; 島田淳一; 前田香織
    研究報告インターネットと運用技術（IOT） 2016-IOT-33 (17) 1 - 1 2188-8787 2016/06
  • 柏崎礼生
    研究報告インターネットと運用技術（IOT） 2016-IOT-33 (16) 1 - 6 2188-8787 2016/05 
    When researchers want bigger budgets than bugets steadily given by parent organization, any grants-in-aid of researches and contracted researches can be a good and attractive choice. In this paper, the author show private overview and results of contracted researches and development with the ministroy of internal affairs and communications (MIC) and consider a management method of researches and developments for a brief period.
  • 広域分散システムの耐障害性を評価する検証プラットフォームの実装と評価
    北口善明; 柏崎礼生; 近堂徹; 市川昊平; 西内一馬; 中川郁夫; 菊池豊
    情報処理学会論文誌 57 (3) 958 - 966 2016/03 [Refereed]
  • Yoshiaki Kitaguchi; Hiroki Kashiwazaki; Tohru Kondo; Kohei Ichikawa; Kazuma Nishiuchi; Ikuo Nakagawa; Yutaka Kikuchi
    情報処理学会論文誌ジャーナル(Web) 57 (3) 958 - 966 1882-7764 2016/03 [Refereed]
     
    It is important to ensure robustness and availability of information infrastructures against disasters and faults. There are many technologies of distribution and redundancy, but most of them do not provide practical validation methods on faults. We propose a platform to evaluate disaster tolerance and fault resistance of distributed systems with an SDN technology emulating artificial disasters and faults. In this paper, we introduce a design and its implementation of the platform, and then also show the effectiveness of the platform through evaluation experiments on it.
  • Ikuo Nakagawa; Hiroki Kashiwazaki; Shinji Shimojo; Kohei Ichikawa; Tohru Kondo; Yoshiaki Kitaguchi; Yutaka Kikuchi; Shigetoshi Yokoyama; Shunji Abe
    5th IIAI International Congress on Advanced Applied Informatics(IIAI-AAI) IEEE Computer Society 867 - 872 2016 [Refereed]
  • Ikuo Nakagawa; Hiroki Kashiwazaki; Shinji Shimojo; Kohei Ichikawa; Tohru Kondo; Yoshiaki Kitaguchi; Yutaka Kikuchi; Shigetoshi Yokoyama; Shunji Abe
    In Proc. of the 5th International Congress on Advanced Applied Informatics (IIAI AAI 2016) IEEE Computer Society 867 - 872 2016 [Refereed]
  • Hiroki Kashiwazaki
    IEEE 39TH ANNUAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE WORKSHOPS (COMPSAC 2015), VOL 3 IEEE 139 - 140 0730-3157 2015 [Invited]
     
    In this paper, several aspects according to Thoman Bittman of Gartner Inc., NIST, Eric Schmidt, Peter F. Drucker and Norbert Wiener are introduced. Even it refers to social responsibility of engineering and technologies. Why we are worrying the cloud computing, especially public cloud? Why it is difficult for us to love the cloud computing? An author wrap it up irresponsibly.
  • Hiroki Kashiwazaki; Seiji Miyanaga; Ichirou Morihara
    インターネットと運用技術シンポジウム2014論文集 2014 93 - 100 2014/11 [Refereed]
     
    A virtualized infrastracture on Osaka University started its service to aggreagte server machines inside the university in March 2010. A hypervisor software of the infrastracture have been VMware ESX 4.0 even now and 105 VMs of 32 systems that using 227 virtual CPU is working on the infrastracture. An university mail systems is one of the systems and this mail system provide 8,450 mail accounts with 38 departments. Although the infrastracture was supposed to be replaced until March 2015, replacement was moved up in September 2014 because of datacenter of the university a.k.a. IT core building was completed of construction in Septmeber 2014 and the current infrastracuture have to be moved to the building. In this paper, authors show the design and construction of new virtualized infrastracture. They also show ICT strategies of cloud computing environment in Osaka University.
  • Hiroki Kashiwazaki; Kazuma Nishiuchi; Yoshiaki Kitaguchi; Kouhei Ichikawa; Tohru Kondou; Ikuo Nakagawa; Yutaka Kikuchi
    情報処理学会研究報告(Web) 2014 (SPT-11) 1 - 8 2014/11 [Refereed]
     
    ICT environment is requisite for our life and business. It is important to ensure toughness and robustness of ICT systems. To ensure toughness and robustness, there are several solution such as redundant configurations, SPoF (single point of failure) less wide-are distribution. Nowaday, we must design our system to endure beyond the scope of assumption of disaster caused by southeaster sea (Tounankai) earthquake. In this paper, authors design and evaluate a platform to evaluate disaster tolerance of ICT system by immitating existing disaster records. To contribute to make countermeasures against disaster and disorder more practical, they propose classification from a point of view of a topology and risk evaluations.
  • Hiroki Kashiwazaki; Seiji Miyanaga; Ichirou Morihara
    IPSJ SIG Technical Reports Information Processing Society of Japan (IPSJ) 2014 (9) 1 - 6 0919-6072 2014/06 
    A virtualized infrastracture in Osaka University started its service to aggreagte server machines inside the university in March 2010. Now, the infrastracture is being enhanced on 2014. One of the purpose of this enforcement is to evaluate perfomances of a software defined storage infrastracture "Virtual SAN" (VSAN) released by VMware Inc. In this paper, we show detail impacts of VSAN, an environment of evaluation and a result of evaluations.
  • Hiroki Kashiwazaki; Seiji Miyanaga; Ichiro Morihara
    研究報告インターネットと運用技術（IOT） 2014-IOT-25 (26) 1 - 6 2014/05 
    A virtualized infrastracture on Osaka University started its service to aggreagte server machines inside the university in March 2010. A hypervisor software of the infrastracture have been VMware ESX 4.0 even now and 105 VMs of 32 systems that using 227 virtual CPU is working on the infrastracture. An university mail systems is one of the systems and this mail system provide 8,450 mail accounts with 38 departments. Although the infrastracture was supposed to be replaced until March 2015, replacement was moved up in September 2014 because of datacenter of the university a.k.a. IT core building will be completed of construction in Septmeber 2014 and the current infrastracuture have to be moved to the building. In this paper, authors show the design and specification of new virtualized infrastracture and its plan of moving.
  • Studies on Adaptive Routing for a Wide-Area Overlay Network System
    Hiroki Kashiwazaki
    Hokkaido University 2014/03 [Refereed]
  • Hiroki Kashiwazaki; Yoshiaki Kitaguchi; Tohru Kondo; Tomohiko Kusuda; Yoshirou Ohnuma; Ikuo Nakagawa; Shunji Abe; Shigetoshi Yokoyama; Shinji Shimojo
    情報処理学会論文誌 55 (3) 1140 - 1150 2014/03 [Refereed]
     
    This paper focuses on distributed storage technologies and backing up data onto remote data centers because of growing sense of disaster recovery for large scale disaster. Now the authors are constructing widely distributed storage environment among Kanazawa University, Hiroshima University, National Institute of Informatics (NII). This environment is characterized by high performance of random I/O access. Because every site can access to same data, IT services can be continued with live migration technology. In this paper, the authors evaluate performances of storage I/O and wide area live migration on this storage environment.
  • Evaluations of widely distributed storage and long distance
    Hiroki Kashiwazaki; Shigetoshi Yokoyama etc
    Information Processing Society of Japan 55 (3) 1 - 11 2014/03 [Refereed]
  • Evaluations of a wide-area distributed storage for a wide-area distributed virtualization infrastracture
    Kashiwazaki Hiroki; Yoshiaki Kitaguchi; Kondo Tohru; Kouhei Ichikawa; Ikuo Nakagawa; Shunji Abe; Shigetoshi Yokoyama; Shinji Shimojo
    Proceedings of NORTH Internet Symposium 2014 33 - 40 1345-0247 2014/02 [Refereed]
     
    The authors have been researching and developing widely distributed storage environment that is characterized by high performance of random I/O access. It is difficult for conventional distributed storage to serve practicable performance for using virtualized machines, especially on live migration. In this paper, the authors show the latest result of performances of I/O on proposed distributed algorithm. A result of 24 thousands km live-migration demonstrated in SC13 (Supercomputer Conference 2013) is also shown.
  • Ikuo Nakagawa; Hiroki Kashiwazaki; Yoshiaki Kitaguchi; Kouhei Ichikawa; Tohru Kondo; Shinji Shimojo
    電子情報通信学会技術研究報告 113 (388) 59 - 64 0913-5685 2014/01 
    We have been involved to research and experiment of Long Distance Live Migration of virtual machines on wide area distributed storage which seems to be a single & shared large POSIX file system. In this paper, we introduce two key technologies in our project; “distributed storage” and “network”. Distributed storage technol- ogy enables to achieve wide area distributed storage by distributing and sharing both meta data and user data in geographically distributed areas. We summarize network technologies in two parts, that is internal network and external network. Especially, the external network technology is important for Long Distance Live Migration so that external hosts (on the Internet, for example) would communicate with a virtual machine even the case where it would be migrated into geographically remote sites. In this paper, we also introduce the result of our experiment.
  • Hiroki Kashiwazaki; Yoshiaki Takai
    2014 38TH ANNUAL IEEE INTERNATIONAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE WORKSHOPS (COMPSACW 2014) IEEE 614 - 619 2014 [Refereed]
     
    Penetration of high definition (HD) video cameras and broadband networks make us be able to stream easily overHD quality movies onto the Internet. When we distribute the movies each other between some facilities, packet losses caused by cross traffic become a serious problem. This paper proposes an adaptive traffic engineering method by using discrete event simulator. This method searches suboptimal paths combination from the enormous combinations of paths determined from the pairs of all source node and all destination node. We use the cloud computing resource to evaluate these combinations rapidly by which we can utilize powerful computing resource. This method is tested in ns-2 simulation on the 11 nodes network. The results show that proposed traffic engineering method has better adaptability and performance in cross traffic avoidance.
  • Hiroki Kashiwazaki
    IPSJ SIG Technical Reports Information Processing Society of Japan (IPSJ) 2013-IOT-22 (13) 1 - 6 0919-6072 2013/07 
    When a virtualization infrastracture which consist of widely distributed institutions connected each other by the Internet, it is so annoying for users to pay charge of VMs because VMs will be migration from one instution to another. The annoyance may make users leave the infrastructure. In this paper, I propose an algorithm of accounting and feedback based on information of usage of computing resource in every institutions.
  • An experimental evaluation of long distance live migration using the wide-area distributed storage
    Yoshiaki Kitaguchi; Tohru Kondo; Hiroki Kashiwazaki; Ikuo Nakagawa; Shinji Shimojo
    信学技報 113 (94, IA2013-7) 37 - 42 2013/06 
    The importance of continued operation of network system has been increasing, large-scale disaster recovery and business continuity planning is required also in research institutions such as universities. We are advancing the establishment of the wide-area distributed storage, which interconnected computing resources in universities and research institutes. It is possible to access the same data from multiple sites by utilizing this wide-area distributed storage, so we can continue a service at another site by a live migration. In this paper, we report an experiment of long distance live migration using the wide-area distributed storage, discuss the effectiveness of this system and future issues.
  • An Example of Constructing an University Virtualization Platform From Small Start.
    Hiroki Kashiwazaki
    Proceedings of NORTH Internet Symposium 2013 19 25 - 32 2013/02 [Refereed]
  • Hiroki Kashiwazaki; Yoshiaki Takai
    情報処理学会論文誌 情報処理学会 54 (1) 404 - 413 1882-7764 2013/01 [Refereed]
     
    By widely spreading of high definition cameras and broadband networks, now we can stream over-HD quality movies onto the Internet easily. When we distribute the movies each other between some facilities, packet losses caused by cross traffic become a serious problem. This paper proposes an adaptive traffic engineering method by using discrete event simulator. This method searches suboptimal paths combination from the enormous combinations of paths determined from the pairs of all source node and all destination node. We use the cloud computing service to evaluate these combinations rapidly, which we can utilize powerful computing resource by. This method is tested in ns-2 simulation on the 11 nodes network. The results show that proposed traffic engineering method has better adaptability and high-speed performance in cross traffic avoidance.
  • Hiroki Kashiwazaki
    インターネットと運用技術シンポジウム2012論文集 2012 94 - 101 2012/12 [Refereed]
     
    In Japanese universities and research institutes, various large-scale academic cloud computing environments were constructed. But in small research institutes, they don't have an access for collavorative use platform among other universities or institutes. So they wouldn't be able to put such huge computing resources to practical use, they would be able to only make the resources obsolete. IaaS (Infrastructure as a Service) is useful even for small institutes. In information systems, it will take a long time to change physical server into virtual server because of not only technical resasons but also management ones. In this paper, I show how to build and change into a virtual environment with cheap budget and rich ideas on grounds of measurements. And then I introduce an inter-cloud computing environment which consists of computers of several institutes. By using this environment, we can realize business continuity plan and disaster recovery.
  • -
    Hiroki Kashiwazaki; Masaharu Munetomo; Yoshiaki Takai
    Proceedings of NORTH Internet Symposium 2012 18 101 - 108 2012/02 [Refereed]
  • Hiroki Kashiwazaki
    SIGUCCS'12 - ACM Proceedings of the SIGUCCS Annual Conference 49 - 52 2012 [Refereed]
     
    Tokyo University of the Arts, which is one of the National Universities of the Arts in Japan, was also affected by the 2011 Tohoku earthquake. Many stored art items in our mu- seums were destroyed and many events were postponed or cancelled. The Art Media Center, which manages the cam- pus networks and the IT policies, used cloud computing ser- vices to overcome these disabilities. For instance, we used the movie streaming service "Ustream" to deliver a movie by the president of our university to substitute for our can- celed graduation and entrance ceremony. We also used the cloud backup storage service " Dropbox Teams" for disaster recovery. By cooperating with other universities, we estab- lished a committee called "RICC" to promote the activities of cloud computing and we are now constructing an exper- imental wide area distributed private inter-cloud computer network. Through explanation of these projects, we will detail for sustainable administration on a University of the Arts in Japan. Copyright © 2012 ACM.
  • Hiroki Kashiwazaki; Satoshi Kobayashi; Shugo Kawai; Norikatsu Ohishi; Yoshiaki Takai
    2012 IEEE/IPSJ 12TH INTERNATIONAL SYMPOSIUM ON APPLICATIONS AND THE INTERNET (SAINT) IEEE 345 - 350 2012 [Refereed]
     
    As the Internet becomes increasingly popular and mature, constant increase of demands and complexities for network traffic has been an big issue. A concentration of large traffic demands to one routing node or one communication line makes a network congestion. This congestion causes increment of total network delay and the total amount of packet losses. These losses grow retransmission of packets, so the congestion goes from bad to worse and the nodes and the lines around the congestion are negatively affected. In order to avoid such congestion and to maintain link quality, various traffic engineering (TE) technologies, which utilize processing ability of routing nodes and the lines, are being researched and developed. However, existing "offline-methods of TE" lack reaction sensitivity and adaptability in topology changes. On the other hand, many of "online-methods of TE" contain single point of failure (SPOF) issues and adjustability issues in a complex topology network. This paper proposes NREI (Network adaptive Routing algorithm for Environmental Intelligence) which is based on an one-way delay algorithm. NREI is developed from the delay time based adaptive routing algorithm which authors proposed in the past. To compare with static routing, NREI is tested in an IP network. Results of the test show that proposed routing algorithm has better adaptability in congested path avoidance and network load balancing.
  • Hiroki Kashiwazaki; Yoshiaki Takai
    インターネットと運用技術シンポジウム2011論文集 2011 1 - 8 2011/12 [Refereed]
     
    Disaster tolerance, autonomy and dependability is strongly desired in broad area movie delivery network such as fixed-point observation. Any conventional methods aren't said to robust especially in large scale natural disaster like major earthquakes. And also it is hard to handle the huge traffic demands and avoid cross traffic and traffic congestion. We think the one reason of this difficulty is that a control plane which constructs routing table and an user plane which transmits packets are congutinated, not set apart in the network routers. A processing power of a control plane is comparatively poorer than the power of high-end processor of late. So conventional router can't realize high quality traffic engineering. Now we propose an adaptive traffic engineering method for small or middle scale multihome network environments which uses a discrete event simulation with cloud computing. This method searches suboptimal paths combination from the enormous combinations of paths determined from the pairs of all source node and all destination node. This method is tested with ns-2 simulator and the results show the high speed performance of searching suboptimal solution.
  • Hiroki Kashiwazaki; Yoshiaki Takai
    Proceedings - 11th IEEE/IPSJ International Symposium on Applications and the Internet, SAINT 2011 480 - 485 2011 [Refereed]
     
    This paper proposes a network routing algorithm REI which has autonomous adaptability to network traffic conditions. When a routing node has some different paths to a given destination, we can evaluate these paths in terms of their latency (delay time) information which will be given from inbound data packets. Based on the path evaluation by latency, every node works as a distributed autonomous agent for adaptive routing. By network simulations to compare with conventional and enhanced OSPFs, we show that the multi-agents based routing algorithm has better adaptability in congested path avoidance and network load balancing. © 2011 IEEE.
  • 離散イベント型シミュレーションを用いたトラフィックエンジニアリング手法の提案
    柏崎礼生; 高井昌彰
    情報処理学会インターネットと運用技術シンポジウム2010論文集（査読付論文） 2010 (14) 35 - 42 2010 [Refereed]
  • KASHIWAZAKI HIROKI; KOBAYASHI SATOSHI; KAWAI SHUGO; OHISHI NORIKATSU; TAKAI YOSHIAKI
    Transactions of Information Processing Society of Japan Information Processing Society of Japan (IPSJ) 49 (3) 1194 - 1203 0387-5806 2008/03 
    As the Internet becomes increasingly popular, constant increase in demand for network traffic has been an issue. In order to avoid traffic congestion and to maintain link quality, various traffic engineering (TE) technologies, which utilize processing ability of a routing node, are being researched and developed. However, existing "offline-method of TE" lacks reaction sensitivity and adaptability in topology changes. On the other hand, "online-method of TE" contains a SPOF (single point of failure) issue and an adjustability issue in a network of complex topology. This paper proposes NREI (...
  • KASHIWAZAKI HIROKI; TAKAI YOSHIAKI
    Transactions of Information Processing Society of Japan Information Processing Society of Japan (IPSJ) 47 (12) 3308 - 3318 0387-5806 2006/12 
    This paper proposes a network routing algorithm REI which has autonomous adaptability to network traffic conditions. When a routing node has some different paths to a given destination, we can evaluate these paths in terms of their latency (delay time) information which will be given from inbound data packets. Based on the path evaluation by latency, every node works as a distributed autonomous agent for adaptive routing. By network simulations to compare with conventional and enhanced OSPFs, we show that the multi-agents based routing algorithm has better adaptability in congested path avo...
  • Kashiwazaki Hiroki; Takai Yoshiaki
    情報科学技術レターズ FIT(電子情報通信学会・情報処理学会)推進委員会 4 99 - 102 2005/08
  • Hiroki Kashiwazaki; Yoshiaki Takai
    Proc. of the 16th IASTED International Conference on Parallel and Distributed Computing And Systems (PDSC2004) 76 - 71 2004/11 [Refereed]
  • Kashiwazaki Hiroki; Takai Yoshiaki
    情報科学技術レターズ FIT(電子情報通信学会・情報処理学会)推進委員会 3 89 - 90 2004/08
  • Hiroki Kashiwazaki; Yoshiaki Takai
    Proc. of IASTED International Conference Networks, Parallel and Distributed Processing, and Applications (IASTED NPDPA2002) 149 - 153 2002/10 [Refereed]
  • H Kashiwazaki; Y Takai
    6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL IV, PROCEEDINGS INT INST INFORMATICS & SYSTEMICS (4) 353 - 356 2002 [Refereed]
     
    This paper introduces a network routing algorithm with autonomous adaptability to changes of network traffic conditions. When a routing node has some different pathes to a certain destination, scores of these available pathes can be evaluated in terms of the latency (delay time) informed back from the destination node. By using such path scores for selection, every node works as a distributed autonomous agent for adaptive routing. In forwarding packets, a routing node decides the next routing node according to the local rules with score evaluation. We have two local rules for adaptive routing: a deterministic rule which selects the next node with shortest latency, and an undeterministic rule which selects the next node at a probability inversely proportional to the latency. A harmonic combination of these two rules leads to the global optimal solution. Our network simulations show that the multiagents-based routing algorithm has good adaptability in congested path avoidance.
  • RICC update on meet46
    Kashiwazaki Hiroki
    ITRC technical report

Conference Activities & Talks

  • A propose of an availability evaluation approach with SDN failure injection testing  [Not invited]
    Hiroki Kashiwazaki
    2nd KISTI-NII workshop  2019/08
  • Distcloud update In 2019 Q2~Q3  [Not invited]
    Hiroki Kashiwazaki
    Asia Pacific Advanced Network 48th meeting  2019/07
  • Hiroki Kashiwazaki
    APAN43: The 43rd Asia-Pacific Advanced Network Meeting  2017/02  New Delhi  APAN
  • Hiroki Kashiwazaki
    APAN43: The 43rd Asia-Pacific Advanced Network Meeting  2017/02  New Delhi  APAN
  • Hiroki Kashiwazaki
    APAN42: The 42nd Asia Pacific Advanced Network Meeting and the 20th Anniversary of APAN  2016/07  Hong Kong  APAN
  • A foresight of DESTCloud  [Invited]
    Hiroki Kashiwazaki
    電子情報通信学会信頼性研究会  2015/10  福岡県福岡市  電子情報通信学会信頼性研究会
     
    It is important to ensure toughness and robustness of infrastructure for network services such as the Internet. Nowaday we, especially people living in the Pacific Ocean side of Japan, must design our system to endure beyond the scope of assumption of disaster caused by southeaster sea (Tounankai) earthquake. In this paper, author shows a "DESTCloud" platform that validates and evaluates disaster tolerances of distributed systems and its future and prospects.
  • 阪大の異常な愛情 〜または私は如何にして心配するのを止めてクラウドを愛するようになったか〜  [Invited]
    柏崎 礼生
    Softbank World 2015  2015/07
  • Hiroki Kashiwazaki
    COMPSAC2015  2015/07  Taichung, Taiwan  IEEE computer society
     
    In this paper, several aspects according to Thoman Bittman of Gartner Inc., NIST, Eric Schmidt, Peter F. Drucker and Norbert Wiener are introduced. Even it refers to social responsibility of engineering and technologies. Why we are worrying the cloud computing, especially public cloud? Why it is difficult for us to love the cloud computing? An author wrap it up irresponsibly.
  • 最新クラウドサービスの実力を徹底追求  [Invited]
    柏崎 礼生
    日経産業新聞フォーラム2014  2014/10
  • Introduction and Evaluations of a Wide Area Distributed Storage System  [Not invited]
    Hiroki Kashiwazaki
    SNIA 11th annual storage developer conference  2014/09
  • Hiroki Kashiwazaki
    第2回シスコアカデミックフォーラム東京  2014/07
  • Hiroki Kashiwazaki
    JAIPA Cloud Conference 2014  2014/03  Tokyo  JAIPA クラウド部会
     
    地域間インタークラウド分科会(RICC)ではSoftware Defined DataCenter を実現するための要素技術研究開発に取り組んでおります。その取り組みの一つとして、ユーザが仮想化ホストのリソース情報を取得することが可能ならどのような利活用が可能となり、またサービス事業者にはどのようなメリットがあるかについて評価をしています。この取り組みと、本分科会のワークショップへのご案内をします。
  • Hiroki Kashiwazaki
    ストレージトレンドセミナー  2013/11  Osaka  SNIA-J
  • Hiroki Kashiwazaki
    SINET及び学認説明会  2013/11  Osaka  国立情報学研究所 SINET利用推進室
  • 教育機関の仮想化基盤の構築と運用の実情  [Invited]
    柏崎 礼生
    ネットワーキングアカデミー・インストラクターズミーティング 2013  2013/08  大阪府  Training Center Japan
     
    計算機資源の仮想化が様々な教育機関で行われていますが、システムの構築や運用には様々な課題があります。大学における仮想化基盤の導入を通して、設計、計測、運用の実情についてご紹介します。
  • 柏崎 礼生
    SS研座談会 「マンガをビッグデータとして扱う」  2013/07  東京都  サイエンティフィック・・システム研究会
     
    マンガそのものは、日本あるいは世界において、大量に流通しているにもかかわらず、そのデータベース化は、全く進捗を見ていない。そこには、マンガそのものの構造上の困難さがあったり、マンガや同人誌に対する図書館の冷遇があったり、あるいは著作権上のややこしさがあったりするのだろうが、いったい何が問題なのかという認識すら、社会的に共有されていないのが現実だ。SS研では、この問題を多角的に議論すべく、まずは「SS研座談会」という小さな形で、この問題を検討する。
  • Hiroki Kashiwazaki
    第1回最新技術動向講演会  2013/07  Tokyo  SNIA-J
     
    大規模災害による危機意識の高まりから災害回復(Disaster Recover:DR)を実現するための技術として遠隔地データセンターでのバックアップや分散ストレージに注目が集まっています。 現在我々は広域分散ストレージ環境を金沢大学、広島大学、NIIを中心に構築し、サービス途絶時間の短い長距離マイグレーションを実現しています。本講演ではこの環境の紹介および評価結果について報告いたします。
  • 東京藝術大学での情報システム仕様書における考察: Webサイト構築とネットワーク更改  [Not invited]
    柏崎 礼生
    平成24年度国立大学法人等情報化発表会  2012/10
  • スモールスタートで始める情報システムの仮想化基盤構築と災害復旧・事業継続計画  [Not invited]
    柏崎 礼生
    平成24年度国立大学法人等情報化発表会  2012/10
  • クラウド: 完全移行の勘所とは  [Invited]
    柏崎 礼生
    サイエンティフィック・システム研究会システム技術分科会  2012/08
  • 東京藝術大学での仮想化基盤構築の勘所などについて  [Invited]
    柏崎 礼生
    2012/06
  • 東京藝術大学におけるCloud Computing利用の推進  [Invited]
    柏崎 礼生
    情報処理学会北海道支部講演会  2012/04
  • Happy Networking and Hacking with Cloud Computing!  [Invited]
    Hiroki Kashiwazaki
    EPNetFaN classroom lecture  2012/02
  • A Horizon for Wide-area Distributed Cloud Computing  [Not invited]
    Hiroki Kashiwazaki
    18th NORTH symposium  2012/02
  • Deploying various effort of Regional Inter-Cloud Committee  [Not invited]
    Hiroki Kashiwazaki
    Academic Cloud Workshop 2012 @ Hiroshima  2012/02
  • 東京芸術大学におけるクラウド活用事例  [Not invited]
    柏崎 礼生
    平成23年度国立大学法人等情報化発表会  2011/11
  • 片方向遅延を用いたトラフィックエンジニアリングアルゴリズムの適応性評価  [Not invited]
    FIT2007  2007
  • 遅延時間情報を利用した自律的な負荷分散を実現する経路制御手法  [Not invited]
    電子情報通信学会NS研究会  2006
  • 自律的な負荷分散と遅延時間増大の抑制を実現する適応的経路制御手法 iREI の性能評価  [Not invited]
    FIT2006  2006
  • 柔軟なトラフィックエンジニアリングを実現する自律分散経路制御手法 REI  [Not invited]
    電子情報通信学会IN研究会  2005
  • REI: An Autonomous Distributed Routing Algorithm'  [Not invited]
    IASTED International Conference on Parallel and Distributed Computing And Systems  2004

Works

  • Slashdot Japan authoring
    2008
  • Regional Internet BackBone
    2004

MISC

Awards & Honors

  • 2020/06 国立情報学研究所 国立情報学研究所長表彰
     新型コロナウイルス感染症対応のためのオンライン会議サービス基盤の整備および利用促進 
    受賞者: 柏崎礼生;込山悠介;坂根栄作;岡本裕子;窪田佳裕;田邊俊治;上村順一;土井光広
  • 2019/03 Information Processing Society of Japan IPSJ Yamashita SIG Research Award
     He loved the cloud, but the cloud did not love him. 
    受賞者: Hiroki Kashiwazaki
  • 2015/07 大阪大学 総長奨励賞
     
    受賞者: 柏崎 礼生
  • 2014/07 Osaka university An encouraging prize of the chancellor of Osaka university
     
    受賞者: Hiroki Kashiwazaki
  • 2013/10 Osaka University 6th Osaka University
     IOTS2012における優秀プレゼンテーション賞に関わる表彰 
    受賞者: Hiroki Kashiwazaki
  • 2012/12 Information Processing Society of Japan Award for Excellent Presentation
     Actuals of Constructing and Management of an University Virtualization Platform From Small Start. japan_society 
    受賞者: Hiroki Kashiwazaki
  • 2004/05 IPSJ Hokkaido division IPSJ Symosium 2004 in Hokkaido, Technical Award
     遅延時間情報ベースの適応的経路制御手法 REI 
    受賞者: Hiroki Kashiwazaki;Yoshiaki Takai
  • 2003 IPSJ Hokkaido division IPSJ Symosium 2003 in Hokkaido, Incentive Award
     広域ネットワークにおける適応性を有する分散ルーティング JPN 
    受賞者: Hiroki Kashiwazaki
  • 2002/07 ISAS SCI The 6th World Multiconference on Systemics, Cybernetics and Informatics Best Paper Award
     Multiagents-Based Adaptive Network Routing Algorithm. 
    受賞者: Hiroki Kashiwazaki;Yoshiaki Takai

Research Grants & Projects

  • Japan Society for the Promotion of Science：Grants-in-Aid for Scientific Research Grant-in-Aid for Early-Career Scientists
    Date (from‐to) : 2019/04 -2022/03 
    Author : 柏崎 礼生
     
    本研究ではサービス利用者やデータ提供源により近い場所に配置されたエッジコンピューティング環境の特質の一つである遅延の低さに注目をし、この特質がゆえにエッジコンピューティング環境では計算機資源の集中と大規模化が困難であり、そのため投資者の自律分散協調によってしか運用品質の向上が期待できないという問題点がある。本研究提案では特に、複数の組織や個人が計算機資源を提供し合うことにより構成されるエッジコンピューティング環境(コミュニティ・エッジ)に着目し、高品質な運用を維持しながらこの環境を持続・拡大させるために、運用品質を根拠としたインセンティブの導入を提案する。提案手法をマルチエージェントシミュレーションを用いて評価するとともに、実際に広く使われるクラウド・コンテナ基盤における提案手法の設計と実装、および広域分散プラットフォームを用いた実環境における評価実験を行い、提案方法の有効性を検証する。また異なるコミュニティ・エッジ間の為替決定方式についても提案と評価を行う。 本研究の1~2年目でマルチエージェントシミュレーションによる評価を実施すると同時に、この評価によって得られた、エッジコンピューティング環境における冗長回線の高可用性における価値の定量的な評価を行い、査読付き国際学会の予稿としてまとめている。3年目は実証実験による評価として、また実際のアプリケーションとしてBeyond 5G環境における透過的な分散処理基盤をターゲットとすることとした。この研究提案は情報通信研究機構の「Beyond 5G シーズ創出型プログラム」として採択され、初歩的な実証評価の設計を行った。この新たな研究開発提案の創出のため当初の目標であった実装評価に遅延が生じたことと、当初成果発表を予定していた国際学会がコロナ禍のため中止となったため、研究期間を一年延長することとしている。
  • 広域分散仮想化環境のためのハイブリッド型トラフィックエンジニアリング手法の研究
    日本学術振興会：科学研究費補助金
    Date (from‐to) : 2014/04 -2017/03 
    Author : 柏崎 礼生
  • 分散システムの耐災害性・耐障害性の検証・評価・反映を行うプラットフォームとビジネスモデルの開発
    総務省：戦略的情報通信研究開発推進事業(SCOPE)重点領域型研究開発(先進的通信アプリケーション開発型)
    Date (from‐to) : 2015/08 -2016/03 
    Author : 柏崎 礼生
  • 分散システムの耐災害性・耐障害性の検証・評価・反映を行うプラットフォームとビジネスモデルの開発
    総務省：戦略的情報通信研究開発推進事業 (SCOPE)
    Date (from‐to) : 2014/08 -2015/03 
    Author : 柏崎 礼生
  • 日本学術振興会：科学研究費補助金
    Date (from‐to) : 2012/04 -2015/03 
    Author : 西村 浩二
  • Emerge the autonomous distributed network routing

Other link

researchmap


Copyright © MEDIA FUSION Co.,Ltd. All rights reserved.