KINDAI UNIVERSITY


*A space between the first name and last name, please enter

MIZUTANI Kimihiro

Profile

FacultyDepartment of Informatics
PositionLecturer
DegreePh.D in Engineering
Commentator Guidehttps://www.kindai.ac.jp/meikan/2330-mizutani-kimihiro.html
URL
Mail
Last Updated :2020/09/02

Education and Career

Education

  •   2005 04  - 2008 03 , Doshisha University, Department of Engineering
  •   2008 04  - 2010 03 , Nara Institute of Science and Technology, Graduate school of Information Science
  •   2014 10  - 2015 09 , Nara Institute of Science and Technology, Graduate school of Information Science

Academic & Professional Experience

  •   2019 04 ,  - 現在, Faculty of Science and Engineering, Department of Informatics, Kindai University
  •   2018 04 ,  - 2019 03 , Assistant Manager, NTT West
  •   2010 04 ,  - 2018 03 , Researcher, Network Innovation Labs, NTT
  •   2008 09 ,  - 2009 03 , Information-technology Promotion Agency,Japan

Research Activities

Research Areas

  • Informatics, Information networks

Research Interests

  • Distributed Networking, IoT, Artificial Intelligence, Overlay Network

Published Papers

  • Reducing Dense Virtual Networks for Fast Embedding, Toru MANO, Takeru INOUE, Kimihiro MIZUTANI, Osamu AKASHI, IEICE Transactions on Communications, IEICE Transactions on Communications, 103(4), - - -, Apr. 2020 , Refereed
  • Increasing Capacity of the Clos Structure for Optical Switching Networks, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Osamu Akashi, Proc. of IEEE Global Communications Conference (GLOBECOM), Proc. of IEEE Global Communications Conference (GLOBECOM), Dec. 2019 , Refereed
  • Measuring Lost Packets with Minimum Counters in Traffic Matrix Estimation., Kohei Watabe, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Osamu Akashi, Kenji Nakagawa, IEICE Transactions, IEICE Transactions, 102-B(1), 76 - 87, 2019 , Refereed
  • On Intelligent Traffic Control For Large Scale Heterogeneous Networks: A Value Matrix Based Deep Learning Approach, Zubair Fadlullah, Fengxiao Tang, Bomin Mao, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE Communications Letters, IEEE Communications Letters, 22(12), 2479 - 2482, Dec. 2018 , Refereed
  • Action Determination Learning Scheme for Driving Support Considering Signal and Traffic Conditions, 水谷后宏, 吉田学, 秦崇洋, 社家一平, 計測自動制御学会論文集, 計測自動制御学会論文集, 54(11), 793‐801(J‐STAGE), 2018
  • Toward Preventive Network Service Management by Neural Networks., Taichi Kawabata, Takashi Kurimoto, Kimihiro Mizutani, 2018 IEEE International Symposium on Local and Metropolitan Area Networks, LANMAN 2018, Washington, DC, USA, June 25-27, 2018, 2018 IEEE International Symposium on Local and Metropolitan Area Networks, LANMAN 2018, Washington, DC, USA, June 25-27, 2018, 125 - 126, 2018 , Refereed
  • A Novel Non-Supervised Deep-Learning-Based Network Traffic Control Method for Software Defined Wireless Networks., Bomin Mao, Fengxiao Tang, Zubair, Md. Fadlullah, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE Wireless Commun., IEEE Wireless Commun., 25(4), 74 - 81, 2018 , Refereed
  • 送信レート制御を行うネットワークにおけるパーフェクトサンプリング, 永田尚志, 井上武, 間野暢, 水谷后宏, 明石修, 電子情報通信学会論文誌 B(Web), 電子情報通信学会論文誌 B(Web), J100-B(12), 1068‐1072 (WEB ONLY), Dec. 2017
  • Towards a Low-Delay Edge Cloud Computing through a Combined Communication and Computation Approach., Tiago Gama Rodrigues, Katsuya Suto, Hiroki Nishiyama, Nei Kato, Kimihiro Mizutani, Takeru Inoue, Osamu Akashi, IEEE 84th Vehicular Technology Conference, VTC Fall 2016, Montreal, QC, Canada, September 18-21, 2016, IEEE 84th Vehicular Technology Conference, VTC Fall 2016, Montreal, QC, Canada, September 18-21, 2016, 1 - 5, 2016 , Refereed
  • Reducing dense virtual networks for fast embedding., Toru Mano, Takeru Inoue, Kimihiro Mizutani, Osamu Akashi, 35th Annual IEEE International Conference on Computer Communications, INFOCOM 2016, San Francisco, CA, USA, April 10-14, 2016, 35th Annual IEEE International Conference on Computer Communications, INFOCOM 2016, San Francisco, CA, USA, April 10-14, 2016, 1 - 9, 2016 , Refereed
  • An efficient framework for data-plane verification with geometric windowing queries., Takeru Inoue,Richard Chen,Toru Mano, Kimihiro Mizutani, Hisashi Nagata, Osamu Akashi, 24th IEEE International Conference on Network Protocols, ICNP 2016, Singapore, November 8-11, 2016, 24th IEEE International Conference on Network Protocols, ICNP 2016, Singapore, November 8-11, 2016, 1 - 10, 2016 , Refereed
  • A mobility-based mode selection technique for fair spatial dissemination of data in multi-channel device-to-device communication., Hideki Kuribayashi, Katsuya Suto,Array, Kimihiro Mizutani, Takeru Inoue, Osamu Akashi, 2016 IEEE International Conference on Communications, ICC 2016, Kuala Lumpur, Malaysia, May 22-27, 2016, 2016 IEEE International Conference on Communications, ICC 2016, Kuala Lumpur, Malaysia, May 22-27, 2016, 1 - 6, 2016 , Refereed
  • Stable Load Balancing with Overlapping ID-space Management in Range-based Structured Overlay Networks, MIZUTANI Kimihiro, INOUE Takeru, MANO Toru, AKASHI Osamu, MATSUURA Satoshi, FUJIKAWA Kazutoshi, Computer Software, Computer Software, 32(3), 3_101 - 3_110, 2015 , Refereed
    Summary:Structured overlay networks that support range queries cannot hash data IDs for load balancing, in order to preserve the total order on the IDs. Since data and queries are not equally distributed on the ID-space without hashing in range-based overlay networks, uneven loads are imposed on the overlay nodes. Existing load balancing techniques for range-based overlay networks distribute the loads by using data reallocation or node migration, which makes the networks very unstable due to heavy data reallocation or frequent churn.<BR>This paper proposes a novel scheme that distributes, fairly, the loads without node migration and with little data reallocation, by sharing some ID-space regions between neighboring nodes. Our &ldquo;overlapping&rdquo; ID-space management scheme derives the optimal overlap based on kernel density estimations; the query loads based on the statistical theory are used to calculate the best overlap regions. This calculation is executed in a distributed manner with no central coordinator. We conduct thorough computer simulations, and show that our scheme alleviates the worst node load by 20&ndash;90 % against existing techniques without node migration and with least data reallocation.
  • TCP Available Bandwidth Estimation Function with Socket API, KUGIMOTO Takeshi, MIYAKE Shigeki, MIZUTANI Kimihiro, Computer Software, Computer Software, 31(3), 3_246 - 3_258, 2014
    Summary:Recently, a lot of streaming services making use of TCP (Transmission Control Protocol) are developed. To keep the QoS in the streaming services highly, it is important to estimate available bandwidth of TCP stream precisely in a short period of time. We propose the real time available bandwidth estimating method. Our proposed method has the two advantages for the conventional bandwidth estimating methods. (1) Our proposed method does not generate the test traffic for the estimation and (2) considers the network conditions such as the jitter and latency. Therefore, our proposed method realizes the real time bandwidth estimation with lower network load and higher precision. Application software can easily use the results obtained by the function through socket API (Application Programming Interface). We implement the method for the TCP protocol stacks and compare the precision with the conventional bandwidth estimation tools. From the experimental results, our method outperforms the tools in the presicion.
  • On the fast-convergence of delay-based load balancing over multipaths for dynamic traffic environments., Meng Li,Array, Nei Kato, Kimihiro Mizutani, Osamu Akashi, Atsushi Takahara, International Conference on Wireless Communications and Signal Processing, WCSP 2013, Hangzhou, China, October 24-26, 2013, International Conference on Wireless Communications and Signal Processing, WCSP 2013, Hangzhou, China, October 24-26, 2013, 1 - 6, 2013 , Refereed
  • Secure resource provisioning across multiple domains., Toru Mano, Kimihiro Mizutani, Osamu Akashi, 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013), Ghent, Belgium, May 27-31, 2013, 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013), Ghent, Belgium, May 27-31, 2013, 1129 - 1134, 2013 , Refereed
  • A Design of Scalable Computing Platform for Continuous Data, MIZUTANI Kimihiro, MANO Toru, AKASHI Osamu, FUKUDA Kensuke, Computer Software, Computer Software, 30(2), 2_101 - 2_118, 2013
    Summary:Recently, many researchers focus on the studies of management and analysis for continuous data such as time series and geographical location data. To manage the large continuous data, the structured overlay network technologies are proposed. In addition, to analyze the large continuous data, MapReduce platforms are proposed. However, it is difficult to analyze the continuous data by MapReduce platforms on overlay. Because general overlay uses hash functions and these hash functions do not preserve the continuousness. In addition, general MapReduce platform generates a lot of communications and synchronous operations in continuous data processing. To handle these problems, we propose the scalable computing platform for continuous data.<BR>Our platform achieves the asynchronous computing and high parallel performance for continuous data analysis. In concrete terms, our platform builds the balanced tree based on SkipList for all nodes. This architecture enables each node to manage its children nodes' states, analysis results, and synchronous operations. In addition, our platform can balance load by gathering all nodes' load information. Therefore, our platform realizes high performance MapReduce computing for continuous data.
  • Dynamic Session Management Based on Reinforcement Learning in Virtual Server Environment, Kimihiro Mizutani, Izumi Koyanagi, Takuji Tachibana, IAENG ICCSA, IAENG ICCSA, 905 - 910, Mar. 2010 , Refereed
  • On Removing Routing Protocol from Future Wireless Networks: A Real-time Deep Learning Approach for Intelligent Traffic Control, Fengxiao Tang, Bomin Mao, Zubair Md. Fadlullah, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE Wireless Communications, IEEE Wireless Communications, 25(1), 154 - 160, Feb. 01 2018 , Refereed
    Summary:Recently, deep learning has appeared as a breakthrough machine learning technique for various areas in computer science as well as other disciplines. However, the application of deep learning for network traffic control in wireless/heterogeneous networks is a relatively new area. With the evolution of wireless networks, efficient network traffic control such as routing methodology in the wireless backbone network appears as a key challenge. This is because the conventional routing protocols do not learn from their previous experiences regarding network abnormalities such as congestion and so forth. Therefore, an intelligent network traffic control method is essential to avoid this problem. In this article, we address this issue and propose a new, real-time deep learning based intelligent network traffic control method, exploiting deep Convolutional Neural Networks (deep CNNs) with uniquely characterized inputs and outputs to represent the considered Wireless Mesh Network (WMN) backbone. Simulation results demonstrate that our proposal achieves significantly lower average delay and packet loss rate compared to those observed with the existing routing methods. We particularly focus on our proposed method's independence from existing routing protocols, which makes it a potential candidate to remove routing protocol(s) from future wired/ wireless networks.
  • A Tensor Based Deep Learning Technique for Intelligent Packet Routing, Bomin Mao, Zubair Md. Fadlullah, Fengxiao Tang, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, 2017 IEEE Global Communications Conference, GLOBECOM 2017 - Proceedings, 2017 IEEE Global Communications Conference, GLOBECOM 2017 - Proceedings, 2018-, 1 - 6, Jan. 10 2018 , Refereed
    Summary:Recently, network operators are confronting the challenge of exploding traffic and more complex network environments due to the increasing number of access terminals having various requirements for delay and package loss rate. However, traditional routing methods based on the maximum or minimum single metric value aim at improving the network quality of only one aspect, which makes them become incapable to deal with the increasingly complicated network traffic. Considering the improvement of deep learning techniques in recent years, in this paper, we propose a smart packet routing strategy with Tensor-based Deep Belief Architectures (TDBAs) that considers multiple parameters of network traffic. For better modeling the data in TDBAs, we use the tensors to represent the units in every layer as well as the weights and biases. The proposed TDBAs can be trained to predict the whole paths for every edge router. Simulation results demonstrate that our proposal outperforms the conventional Open Shortest Path First (OSPF) protocol in terms of overall packet loss rate and average delay per hop.
  • Fast packet classification algorithm for network-wide forwarding behaviors, Takeru Inoue, Toru Mano, Kimihiro Mizutani, Shin-ichi Minato, Osamu Akashi, Computer Communications, Computer Communications, 116, 101 - 117, Jan. 01 2018 , Refereed
    Summary:Packet classification has been a key technology to quickly identify an action to be taken on a packet at a switch. Several advanced applications, most of which have been introduced recently with the advent of software-defined networking, commonly require the identification of a combination of switch actions, i.e., the network-wide forwarding behavior of a packet. Conventional classification methods, however, fail to well support network-wide behaviors, since the search space is partitioned in convoluted manner due to the complexity posed by the combinations possible. This paper proposes a novel packet classification method that supports the fast determination of network-wide forwarding behaviors. To avoid the inefficiencies of existing methods, which are revealed for the first time by our research, we base our method on a compressed data structure named the multi-valued decision diagram. On the solid foundation of decision diagrams, several algorithms are introduced with thorough theoretical analyses, and the construction process and the classification performance are highly optimized for the new classification problem. Experiments on real network datasets show that our method identifies the network-wide forwarding behaviors at the line basic rate, e.g., 10 Mpps, on a single CPU core with only tens of MB of memory.
  • An Efficient Framework for Data-Plane Verification With Geometric Windowing Queries, Takeru Inoue, Richard Chen, Toru Mano, Kimihiro Mizutani, Hisashi Nagata, Osamu Akashi, IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 14(4), 1113 - 1127, Dec. 2017 , Refereed
    Summary:Modern networks have complex configurations to provide advanced functions. Network softwarization, a promising new movement in the networking community, could make networks more complexly configured due to the nature of software. Since these complexities make the networks error-prone, network verification is attracting attention as a key technology to detect inconsistencies between a configuration and an operational policy. Existing verifiers are, unfortunately, either inefficient or incomplete (operational policies are not rigorously checked). This paper presents a novel framework of data-plane verification. So as to efficiently manage the large search space defined by packet headers, our framework formalizes the consistency check by applying simple set operations defined in a small quotient space of packet header. This paper also reveals that the two spaces can be connected via the windowing query in computational geometry. Two windowing algorithms are proposed and backed by solid theoretical analyses. Experiments on real network datasets show that our framework with the windowing algorithms is surprisingly fast; when verifying policy compliance in a real network with thousands of switches, our framework reduces the verification time of all-pairs reachability from ten hours to ten minutes.
  • Routing or Computing? The Paradigm Shift Towards Intelligent Computer Network Packet Transmission Based on Deep Learning, Bomin Mao, Zubair Md. Fadlullah, Fengxiao Tang, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE TRANSACTIONS ON COMPUTERS, IEEE TRANSACTIONS ON COMPUTERS, 66(11), 1946 - 1960, Nov. 2017 , Refereed
    Summary:Recent years, Software Defined Routers (SDRs) (programmable routers) have emerged as a viable solution to provide a cost-effective packet processing platform with easy extensibility and programmability. Multi-core platforms significantly promote SDRs' parallel computing capacities, enabling them to adopt artificial intelligent techniques, i.e., deep learning, to manage routing paths. In this paper, we explore new opportunities in packet processing with deep learning to inexpensively shift the computing needs from rule-based route computation to deep learning based route estimation for high-throughput packet processing. Even though deep learning techniques have been extensively exploited in various computing areas, researchers have, to date, not been able to effectively utilize deep learning based route computation for high-speed core networks. We envision a supervised deep learning system to construct the routing tables and show how the proposed method can be integrated with programmable routers using both Central Processing Units (CPUs) and Graphics Processing Units (GPUs). We demonstrate how our uniquely characterized input and output traffic patterns can enhance the route computation of the deep learning based SDRs through both analysis and extensive computer simulations. In particular, the simulation results demonstrate that our proposal outperforms the benchmark method in terms of delay, throughput, and signaling overhead.
  • THE DEEP LEARNING VISION FOR HETEROGENEOUS NETWORK TRAFFIC CONTROL: PROPOSAL, CHALLENGES, AND FUTURE PERSPECTIVE, Nei Kato, Zubair Md. Fadlullah, Bomin Mao, Fengxiao Tang, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE WIRELESS COMMUNICATIONS, IEEE WIRELESS COMMUNICATIONS, 24(3), 146 - 153, Jun. 2017 , Refereed
    Summary:Recently, deep learning, an emerging machine learning technique, is garnering a lot of research attention in several computer science areas. However, to the best of our knowledge, its application to improve heterogeneous network traffic control (which is an important and challenging area by its own merit) has yet to appear because of the difficult challenge in characterizing the appropriate input and output patterns for a deep learning system to correctly reflect the highly dynamic nature of large-scale heterogeneous networks. In this vein, in this article, we propose appropriate input and output characterizations of heterogeneous network traffic and propose a supervised deep neural network system. We describe how our proposed system works and how it differs from traditional neural networks. Also, preliminary results are reported that demonstrate the encouraging performance of our proposed deep learning system compared to a benchmark routing strategy (Open Shortest Path First (OSPF)) in terms of significantly better signaling overhead, throughput, and delay.
  • A Proposal of an Efficient Traffic Matrix Estimation under Packet Drops, Kohei Watabe, Toru Mano, Kimihiro Mizutani, Osamu Akashi, Kenji Nakagawa, Takeru Inoue, 2017 IEEE 37TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS 2017), 2017 IEEE 37TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS (ICDCS 2017), 2636 - 2637, 2017 , Refereed
    Summary:Traffic matrix (TM) estimation has been extensively studied for decades. Although conventional estimation techniques assume that traffic volumes are unchanged between origins and destinations, packets are often discarded on a path due to traffic burstiness, silent failures, etc. This paper proposes a novel TM estimation method that works correctly even under packet drops. The method is established on a Boolean fault localization technique; the technique requires fewer counters though it only determines whether each link is healthy. This paper extends the Boolean technique so as to deal with traffic volumes with error bounds just by a small number of counters. Along with submodular optimization for the minimum counter placement, we evaluate our method with real network datasets.
  • State-of-the-Art Deep Learning: Evolving Machine Intelligence Toward Tomorrow's Intelligent Network Traffic Control Systems, Zubair Md. Fadlullah, Fengxiao Tang, Bomin Mao, Nei Kato, Osamu Akashi, Takeru Inoue, Kimihiro Mizutani, IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, 19(4), 2432 - 2455, 2017 , Refereed
    Summary:Currently, the network traffic control systems are mainly composed of the Internet core and wired/wireless heterogeneous backbone networks. Recently, these packet-switched systems are experiencing an explosive network traffic growth due to the rapid development of communication technologies. The existing network policies are not sophisticated enough to cope with the continually varying network conditions arising from the tremendous traffic growth. Deep learning, with the recent breakthrough in the machine learning/intelligence area, appears to be a viable approach for the network operators to configure and manage their networks in a more intelligent and autonomous fashion. While deep learning has received a significant research attention in a number of other domains such as computer vision, speech recognition, robotics, and so forth, its applications in network traffic control systems are relatively recent and garnered rather little attention. In this paper, we address this point and indicate the necessity of surveying the scattered works on deep learning applications for various network traffic control aspects. In this vein, we provide an overview of the state-of-the-art deep learning architectures and algorithms relevant to the network traffic control systems. Also, we discuss the deep learning enablers for network systems. In addition, we discuss, in detail, a new use case, i.e., deep learning based intelligent routing. We demonstrate the effectiveness of the deep learning-based routing approach in contrast with the conventional routing strategy. Furthermore, we discuss a number of open research issues, which researchers may find useful in the future.
  • Efficient Virtual Network Optimization Across Multiple Domains Without Revealing Private Information, Toru Mano, Takeru Inoue, Dai Ikarashi, Koki Hamada, Kimihiro Mizutani, Osamu Akashi, IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, 13(3), 477 - 488, Sep. 2016 , Refereed
    Summary:Building optimal virtual networks across multiple domains is an essential technology for offering flexible network services. However, existing research is founded on an unrealistic assumption: providers will share their private information including resource costs. Providers, as well known, never actually do that so as to remain competitive. Secure multi-party computation, a computational technique based on cryptography, can be used to secure optimization, but it is too time consuming. This paper presents a novel method that can optimize virtual networks built over multiple domains efficiently without revealing any private information. Our method employs secure multi-party computation only for masking sensitive values; it can optimize virtual networks under limited information without applying any time-consuming techniques. It is solidly based on the theory of optimality and is assured of finding reasonably optimal solutions. Experiments show that our method is fast and optimal in practice, even though it conceals private information; it finds near optimal solutions in just a few minutes for large virtual networks with tens of nodes. This is the first work that can be implemented in practice for building optimal virtual networks across multiple domains.
  • Living Will for Resilient Structured Overlay Networks, Kimihiro Mizutani, Takeru Inoue, Toru Mano, Osamu Akashi, Satoshi Matsuura, Kazutoshi Fujikawa, IEICE TRANSACTIONS ON COMMUNICATIONS, IEICE TRANSACTIONS ON COMMUNICATIONS, E99B(4), 830 - 840, Apr. 2016 , Refereed
    Summary:The routing efficiency of structured overlay networks depends on the consistency of pointers between nodes, where a pointer maps a node identifier to the corresponding address. This consistency can, however, break temporarily when some overlay nodes fail, since it takes time to repair the broken pointers in a distributed manner. Conventional solutions utilize "backpointers" to quickly discover any failure among the pointing nodes, which allow them to fix the pointers in a short time. Overlay nodes are, however, required to maintain backpointers for every pointing node, which incurs significant memory and consistency check overhead. This paper proposes a novel light-weight protocol; an overlay node gives a "living will" containing its acquaintances (backpointers) only to its successor, thus other nodes are freed from the need to maintain it. Our carefully-designed protocol guarantees that all acquaintances are registered via the living will, even in the presence of churn, and the successor notifies the acquaintances for the deceased. Even if the successor passes away and the living will is lost, the successor to the successor can identify the acquaintances with a high success ratio. Simulations show that our protocol greatly reduces memory overhead as well as the detection time for node failure with the cost being a slight increase in messaging load.
  • A Geometric Windowing Algorithm in Network Data-plane Verification, Richard Chen, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Hisashi Nagata, Osamu Akashi, PROCEEDINGS 2016 IEEE 36TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS ICDCS 2016, PROCEEDINGS 2016 IEEE 36TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS ICDCS 2016, 743 - 744, 2016 , Refereed
    Summary:Network verification is attracting attention as a key technology to detect configuration errors before deploying the network. In verification, a set of packets to be inspected is usually specified by a window - a multi-dimensional rectangle defined by packet header fields (e.g., address prefixes and port ranges). Network operators have to know the forwarding behaviors of packets inside the window; this can be regarded as the windowing query problem in computation geometry. This paper proposes a novel windowing algorithm for network verification. Unlike existing windowing algorithms, our algorithm runs on a compressed data structure, because the search space has to be represented in a compressed form due to the space complexity.
  • Virtual Network Embedding across Multiple Domains with Secure Multi-Party Computation, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Osamu Akashi, IEICE TRANSACTIONS ON COMMUNICATIONS, IEICE TRANSACTIONS ON COMMUNICATIONS, E98B(3), 437 - 448, Mar. 2015 , Refereed
    Summary:Network virtualization is one of the promising technologies that can increase flexibility, diversity, and manageability of networks. Building optimal virtual networks across multiple domains is getting much attention, but existing studies were based on an unrealistic assumption, that is, providers' private information can be disclosed; as is well known, providers never actually do that. In this paper, we propose a new method that solves this multi-domain problem without revealing providers' private information. Our method uses an advanced secure computation technique called multi-party computation (MPC). Although MPC enables existing unsecured methods to optimize virtual networks securely, it requires very large time to finish the optimization due to the MPC's complex distributed protocols. Our method, in contrast, is designed to involve only a small number of MPC operations to find the optimal solution, and it allows providers to execute a large part of the optimization process independently without heavy distributed protocols. Evaluation results show that our method is faster than an existing method enhanced with MPC by several orders of magnitude. We also unveil that our method has the same level of embedding cost.
  • Testaments for Resilient Structured Overlay Networks, Kimihiro Mizutani, Takeru Inoue, Toru Mano, Osamu Akashi, Satoshi Matsuura, Kazutoshi Fujikawa, 2015 21ST ASIA-PACIFIC CONFERENCE ON COMMUNICATIONS (APCC), 2015 21ST ASIA-PACIFIC CONFERENCE ON COMMUNICATIONS (APCC), 514 - 518, 2015 , Refereed
    Summary:The routing efficiency of structured overlay networks depends on the consistency of pointers between nodes. This consistency can, however, break temporarily when some overlay nodes fail, since it takes time to repair the broken pointers in a distributed manner. Conventional solutions utilize "backpointers" to quickly know the failure among the pointing nodes, which allows them to fix the pointers in a short time. Overlay nodes are, however, required to maintain backpointers for every pointing node which incurs significant consistency check and memory overheads. This paper proposes a novel light-weight protocol; an overlay node gives a "testament" containing its acquaintances (backpointers) only to its successor (i.e., clockwise closest node), and other nodes are freed from maintaining it. Our carefully-designed protocol guarantees that all acquaintances are registered with the testament even in the presence of churn, and the successor notifies the acquaintances for the deceased. Even if the successor passes away and the testament is lost, the successor of the successor can identify the acquaintances at a high success ratio. Simulations show that our protocol greatly reduces the mean time to repair (MTTR) and memory overheads while messaging cost increases.
  • An Overlay-Based Data Mining Architecture Tolerant to Physical Network Disruptions, Katsuya Suto, Hiroki Nishiyama, Nei Kato, Kimihiro Mizutani, Osamu Akashi, Atsushi Takahara, IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, 2(3), 292 - 301, Jul. 2014 , Refereed
    Summary:Management scheme for highly scalable big data mining has not been well studied in spite of the fact that big data mining provides many valuable and important information for us. An overlay based parallel data mining architecture, which executes fully distributed data management and processing by employing the overlay network, can achieve high scalability. However, the overlay-based parallel mining architecture is not capable of providing data mining services in case of the physical network disruption that is caused by router/communication line breakdowns because numerous nodes are removed from the overlay network. To cope with this issue, this paper proposes an overlay network construction scheme based on node location in physical network, and a distributed task allocation scheme using overlay network technology. The numerical analysis indicates that the proposed schemes considerably outperform the conventional schemes in terms of service availability against physical network disruption.
  • Rethinking Packet Classification for Global Network View of Software-Defined Networking, Takeru Inoue, Toru Mano, Kimihiro Mizutani, Shin-ichi Minato, Osamu Akashi, 2014 IEEE 22ND INTERNATIONAL CONFERENCE ON NETWORK PROTOCOLS (ICNP), 2014 IEEE 22ND INTERNATIONAL CONFERENCE ON NETWORK PROTOCOLS (ICNP), 296 - 307, 2014 , Refereed
    Summary:In software-defined networking, applications are allowed to access a global view of the network so as to provide sophisticated functionalities, such as quality-oriented service delivery, automatic fault localization, and network verification. All of these functionalities commonly rely on a well-studied technology, packet classification. Unlike the conventional classification problem to search for the action taken at a single switch, the global network view requires to identify the network-wide behavior of the packet, which is defined as a combination of switch actions. Conventional classification methods, however, fail to well support network-wide behaviors, since the search space is complicatedly partitioned due to the combinations. This paper proposes a novel packet classification method that efficiently supports network-wide packet behaviors. Our method utilizes a compressed data structure named the multi-valued decision diagram, allowing it to manipulate the complex search space with several algorithms. Through detailed analysis, we optimize the classification performance as well as the construction of decision diagrams. Experiments with real network datasets show that our method identifies the packet behavior at 20.1 Mpps on a single CPU core with only 8.4 MB memory; by contrast, conventional methods failed to work even with 16 GB memory. We believe that our method is essential for realizing advanced applications that can fully leverage the potential of software-defined networking.
  • Efficient Virtual Network Optimization across Multiple Domains without Revealing Private Information, Toru Mano, Takeru Inoue, Dai Ikarashi, Koki Hamada, Kimihiro Mizutani, Osamu Akashi, 2014 23RD INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND NETWORKS (ICCCN), 2014 23RD INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND NETWORKS (ICCCN), 1 - 8, 2014 , Refereed
    Summary:Building optimal virtual networks across multiple domains is an essential technology to offer flexible network services. However, existing research is founded on an unrealistic assumption; providers will share their private information including resource costs. Providers, as is well known, never actually do that to remain competitive. Technically, secure multiparty computation, which is a computational technique based on the cryptography, can be used to secure optimization, but it is too time-consuming. This paper presents a novel method to optimize virtual networks built over multiple domains, with great efficiency but without revealing any private information. Our method employs secure multi-party computation but only for masking sensitive values; it can optimize virtual networks under limited information without any time-consuming technique. It is solidly based on the theory of optimality, and is assured of finding reasonably optimal solutions. Experiments show that our method is fast and optimal in practice even concealing private information; it finds nearly optimal solutions in just a few minutes for large virtual networks with tens of nodes. This is the first work that can be implemented in practice for building optimal virtual networks across multiple domains.
  • Efficient Virtual Network Optimization across Multiple Domains without Revealing Private Information, Toru Mano, Takeru Inoue, Dai Ikarashi, Koki Hamada, Kimihiro Mizutani, Osamu Akashi, 2014 23RD INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND NETWORKS (ICCCN), 2014 23RD INTERNATIONAL CONFERENCE ON COMPUTER COMMUNICATION AND NETWORKS (ICCCN), 13(3), 477 - 488, 2014 , Refereed
    Summary:Building optimal virtual networks across multiple domains is an essential technology to offer flexible network services. However, existing research is founded on an unrealistic assumption; providers will share their private information including resource costs. Providers, as is well known, never actually do that to remain competitive. Technically, secure multiparty computation, which is a computational technique based on the cryptography, can be used to secure optimization, but it is too time-consuming. This paper presents a novel method to optimize virtual networks built over multiple domains, with great efficiency but without revealing any private information. Our method employs secure multi-party computation but only for masking sensitive values; it can optimize virtual networks under limited information without any time-consuming technique. It is solidly based on the theory of optimality, and is assured of finding reasonably optimal solutions. Experiments show that our method is fast and optimal in practice even concealing private information; it finds nearly optimal solutions in just a few minutes for large virtual networks with tens of nodes. This is the first work that can be implemented in practice for building optimal virtual networks across multiple domains.
  • MARIF: Multiple Queries Look-Up Architecture Using Range Information Feedback in a DHT Network, Kimihiro Mizutani, Toru Mano, Osamu Akashi, Kensuke Fukuda, IEICE TRANSACTIONS ON COMMUNICATIONS, IEICE TRANSACTIONS ON COMMUNICATIONS, E96B(7), 1680 - 1690, Jul. 2013 , Refereed
    Summary:In DHT network, a node can get/put a requested data by only log N look-up steps. However, conventional DHT network only supports single query look-up to search data. From the reason, each node in a DHT network must execute look-up process for each query even if a large number of put and get operations are executed. Therefore, this results in high network load in massive data management such as Map Reduce, sensor network, and web information. To address the problem, we propose multiple queries look-up architecture using range information feedback (MARIF). MARIF extends the conventional KBR protocol to supports range information that is a scope of ID space a node keeps. When a source node receives range information from a destination node, the source node checks all queries in the range information and forwards queries matching the range information to the destination node directly. This effectively reduces the number of look-up queries and the network load for the IP network. In addition, MARIF can be implemented into conventional DHT networks and can easily be combined to effective DHT routing algorithms such as Chord, Kademlia, Pastry, and one-hop DHT. In evaluation, we implement MARIF into three DHT networks and compare its performance with that of conventional query bundling mechanisms based on the KBR protocol. The results show that MARIF reduces by up to 40% the total number of forwarding queries to put data compared with other mechanisms. In addition, MARIF saves the number of forwarding queries per look-up process by up to 85% compared to other mechanisms with low bundling overhead.
  • Streaming Server Management Scheme for Reducing Power Consumption, Kimihiro Mizutani, Toru Mano, Osamu Akashi, Tetsuo Kawano, Hiroshi Shimizu, 2012 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2012 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 3548 - 3553, 2012 , Refereed
    Summary:In this paper, we propose an efficient video streaming server management scheme for streaming service that minimizes power consumption while satisfying client's requests. In conventional schemes, the power minimizing policy is calculated by means of simple dynamic programming or linear programing systems. However, these schemes may suffer changing the power minimizing policy substantially when the client's requests change even slightly. This increases the cost of changing streaming servers. On the other hand, our approach minimizes power consumption while satisfying client's requests by considering the number of times the policy is changed. To reduce the number of changes, our scheme uses dynamic programming by considering trends in client's requests and the stream length for each server. Evaluations shows that our scheme reduces power consumption by up to 35%-45% of conventional schemes.
  • Efficient Query Bundling Mechanism in a DHT Network, Kimihiro Mizutani, Toru Mano, Osamu Akashi, Kensuke Fukuda, 2012 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2012 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2695 - 2700, 2012 , Refereed
    Summary:A distributed hash table (DHT) network can be used for many distributed services and systems. In DHT networks, it takes log N look-up steps to search for required data where N is the number of nodes. However, the look-up process is redundant in the IP network because each look-up process generates a lot of communication among nodes. In massive data management such as sensor and web information management, this results in high network load even if each the search process takes only log N look-up steps. To solve this problem, we propose an efficient query bundling mechanism that makes it possible to bundle multiple queries by using range information. Range information consists of ID space information kept by a node. When a source node receives range information from a destination node, the source node matches all queries for the range information and forwards queries matching the range information to the destination node directly. This effectively reduces the number of look-up processes and the network load for the IP network. In addition, our mechanism can be implemented into conventional DHT networks and can easily be combined to effective DHT routing algorithms such as Chord, Kademlia, and Pastry. In evaluation, we implement our mechanism into DHT networks and compare its performance with that of conventional query bundling mechanisms. The results show that our mechanism reduces by up to 75% the total number of forwarding operations to put data compared with other mechanisms. In addition, our mechanism realizes the reduction of the number of forwarding operations per look-up process by up to 85% compared to other mechanisms.
  • A Dynamic Flow Control Mechanism Based on a Hierarchical Feedback Model for Data Center Networks, Kimihiro Mizutani, Osamu Akashi, Atsushi Terauchi, Takeshi Kugimoto, Mitsuru Maruyama, 2012 IEEE NETWORK OPERATIONS AND MANAGEMENT SYMPOSIUM (NOMS), 2012 IEEE NETWORK OPERATIONS AND MANAGEMENT SYMPOSIUM (NOMS), 599 - 602, 2012 , Refereed
    Summary:Many studies have focused on developing Transmission Control Protocol (TCP) flow control mechanisms to enable more effective use of bandwidth in data center networks. However, the studies cannot achieve total optimization of bandwidth use for data center networks. This is because they do not take into account the path design of TCP flows and the hierarchical structure in data center networks through which TCP flows. To address this issue, we propose a TCP flow control mechanism for effectively using bandwidth by dynamically controlling paths through which TCP flows. To achieve this dynamic control, a hierarchical feedback model is used to obtain an optimal path establishment policy. With the model, all network components exchange their resource states with each other when each network component establishes a TCP flow. This enables each network component to learn the best TCP flow path for effective bandwidth use. An evaluation process shows that the mechanism achieves more effective bandwidth use and more efficient flow control than the conventional TCP flow control.
  • An implementation and its evaluation of a framework for managing states of nodes among structured overlay networks, Kimihiro Mizutani, Satoshi Matsuura, Shinichi Doi, Kazutoshi Fujikawa, Hideki Sunahara, 6th International Conference on Networking and Services, ICNS 2010, Includes LMPCNA 2010; INTENSIVE 2010, 6th International Conference on Networking and Services, ICNS 2010, Includes LMPCNA 2010; INTENSIVE 2010, 282 - 287, 2010 , Refereed
    Summary:Structured overlay networks are widely used as platforms for distributed systems. However, in a high churn situation, it is highly costly to maintain the structure and service availability of an overlay network because there is a limitation in churn handling. Several cooperation mechanisms among overlay networks are proposed to improve service availability of overlay networks in a high-churn situation. Several cooperation mechanisms are proposed among overlay networks. The cooperation mechanisms can group structured overlay networks to simplify routing functions. However, the existing mechanisms do not consider managing churn information among overlay networks. In this paper, we propose a new framework that can share states of the nodes and spread churn information to neighbor nodes among different overlay networks. Using our framework, we are able to maintain each overlay network effectively and improve the reachability of messages under churn. We evaluate our framework with Chord, Kademlia, and Pastry. Our framework improves the detection time of churn by about 30% and decreases the stabilization messages by about 30-40%. We discuss combinatorial effects of the sharing state of the nodes among different overlay networks. © 2010 IEEE.

Misc

  • Increasing Capacity of the Clos Structure for Efficient Nonblocking Networks, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Osamu Akashi, 118, 466, 25, 30,   2019 03 , https://www.ieice.org/ken/paper/20190304j1l6/
  • トマトの生育情報を用いた収穫日予測手法の検討, 水谷后宏, 東田光裕, 増田貴大, 岡本朋子, 電子情報通信学会大会講演論文集(CD-ROM), 2018, ROMBUNNO.B‐18‐22,   2018 08 28 , http://jglobal.jst.go.jp/public/201802255789925299
  • 深層強化学習を用いた一般道における自動車の速度調整システム, 水谷后宏, 吉田学, 秦崇弘, 社家一平, 電子情報通信学会大会講演論文集(CD-ROM), 2018, ROMBUNNO.A‐14‐14,   2018 03 06 , http://jglobal.jst.go.jp/public/201802251739539321
  • Network Reduction Method that Minimizes Total Capacity Change for Virtual Network Embedding Acceleration, Toru Mano, Takeru Inoue, Kimihiro Mizutani, Shin-ichi Minato, Osamu Akashi, 117, 129, 13, 18,   2017 07 , http://www.ieice.org/ken/paper/20170719Dbvz/
  • ネットワーク故障におけるパスの統合・分割による故障単位の推定, 堤陽祐, 渡部康平, 井上武, 水谷后宏, 間野暢, 明石修, 中川健治, 電子情報通信学会大会講演論文集(CD-ROM), 2017, ROMBUNNO.B‐7‐62,   2017 03 07 , http://jglobal.jst.go.jp/public/201702251701646242
  • On Efficient Framework for Data-plane Verification with Geometric Windowing Queries, Takeru Inoue, Toru Mano, Kimihiro Mizutani, Hisashi Nagata, Osamu Akashi, IEICE Technical Report, 116, 251, 25, 30,   2016 10 , http://ci.nii.ac.jp/naid/40020991526
  • パーフェクトシミュレーション高速化のための遷移先割り当てに関する一検討, 永田尚志, 井上武, 水谷后宏, 間野暢, 明石修, 電子情報通信学会大会講演論文集(CD-ROM), 2016, ROMBUNNO.B‐7‐23,   2016 09 06 , http://jglobal.jst.go.jp/public/201602285804484395
  • An Analysis Method of Network Troubles with Path Identificationz, 渡部 康平, 井上 武, 水谷 后宏, 間野 暢, 明石 修, 中川 健治, 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 115, 484, 135, 140,   2016 03 03 , http://ci.nii.ac.jp/naid/40020791215
  • Interactive Analysis of Large-scale Networks based on Integration of Multiple-layer Information, 明石 修, 水谷 后宏, 間野 暢, 井上 武, 福田 健介, 漆谷 重雄, 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 115, 481, 215, 220,   2016 03 03 , http://ci.nii.ac.jp/naid/40020791228
  • Interactive Analysis of Large-scale Networks based on Integration of Multiple-layer Information, 明石 修, 水谷 后宏, 間野 暢, 井上 武, 福田 健介, 漆谷 重雄, 電子情報通信学会技術研究報告 = IEICE technical report : 信学技報, 115, 482, 215, 220,   2016 03 03 , http://ci.nii.ac.jp/naid/40020791971
  • Distributed Time Series Data Management and Analysis, Kimihiro Mizutani, Takeru Inoue, Toru Mano, Hisashi Nagata, Osamu Akashi, NTT Technical Review, 16, 3, 1, 6,   2016 03 , https://www.ntt-review.jp/archive/ntttechnical.php?contents=ntr201603fa5.html
  • Interactive Analysis of Large-scale Networks based on Integration of Multiple-layer Information, 明石修, 水谷后宏, 間野暢, 井上武, 福田健介, 漆谷重雄, 情報処理学会研究報告(Web), 2016, IOT-32, 2016-IOT-32(35) (WEB ONLY),   2016 02 25 , http://jglobal.jst.go.jp/public/201602200004054411
  • An Analysis Method of Network Troubles with Path Identification, 渡部康平, 井上武, 水谷后宏, 間野暢, 明石修, 中川健治, 電子情報通信学会技術研究報告, 115, 484(IN2015 108-153), 135‐140,   2016 02 25 , http://jglobal.jst.go.jp/public/201602203327666643
  • Debuggable reinforcement learning for server resource management, 水谷后宏, 井上武, 間野暢, 明石修, 電子情報通信学会技術研究報告, 115, 409(ICM2015 26-40), 31‐36,   2016 01 14 , http://jglobal.jst.go.jp/public/201602208344568216
  • 新世代ネットワークと革新的サービス 時系列データの分散管理・解析技術, 水谷后宏, 井上武, 間野暢, 永田尚志, 明石修, NTT技術ジャーナル, 28, 1, 30, 34,   2016 01 01 , http://jglobal.jst.go.jp/public/201602212686619405
  • Efficient Network Policy Checking with Multi-dimensional Graph Traversal Algorithm, Chen Richard, Takeru Inoue, Toru Mano, Kimihiro Mizutani, Hisashi Nagata, Osamu Akashi, IEICE Technical Report, 115, 256, 25, 30,   2015 10 , http://ci.nii.ac.jp/naid/40020651210
  • Packet Classification for Global Network View of SDN with MDDs, INOUE T, MANO T, MIZUTANI K, MINATO S, AKASHI O, IEICE technical report. Information networks, 114, 207, 1, 6,   2014 09 11 , http://ci.nii.ac.jp/naid/110009950830
    Summary:In software-defined networking, applications are allowed to access a global view of the network so as to provide sophisticated functionalities. They commonly rely on packet classification, but unlike the conventional classification problem to search for the action taken at a single switch, the global network view requires to identify the network-wide behavior of the packet. Conventional classification methods, however, fail to well support network-wide behaviors, since the search space is complicatedly partitioned. This paper proposes a novel packet classification method that efficiently supports network-wide packet behaviors. Our method utilizes a compressed data structure named the multi-valued decision diagram, allowing it to manipulate the complex search space with several algorithms. Experiments with real network datasets show that our method identifies the packet behavior at 20.1 Mpps on a single CPU core with only 8.4 MB memory.
  • Virtual Network Optimization across Multiple Domains without Revealing Private Information, MANO Toru, INOUE Takeru, IKARASHI Dai, HAMADA Koki, MIZUTANI Kimihiro, AKASHI Osamu, IEICE technical report. Information networks, 114, 139, 83, 88,   2014 07 17 , http://ci.nii.ac.jp/naid/110009946990
    Summary:Building optimal virtual networks across multiple domains is an essential technology to offer flexible network services. However, existing research unrealistically assumes that providers will share their private information. Providers, as is well known, never actually do that to remain competitive. Technically, secure multi-party computation, which is a cryptographic tool, can be used to secure optimization, but it is too time-consuming. This paper presents a novel method to optimize virtual networks built over multiple domains, with great efficiency but without revealing any private information. Our method employs secure multi-party computation but only for masking sensitive values to evade time-consuming computation. It is solidly based on the theory of optimality, and is assured of finding reasonably optimal solutions. Experiments show that our method finds nearly optimal solutions in just a few minutes for virtual network with tens of nodes.
  • Packet Classification for Global Network View of Software-Defined Networking, Takeru Inoue, Toru Mano, Kimihiro Mizutani, Shin-ichi Minato, Osamu Akashi, TCS Technical Reports, Division of Computer Science, Hokkaido University, A, 14, 74,   2014 07 , http://www-alg.ist.hokudai.ac.jp/~thomas/TCSTR/tcstr_14_74/tcstr_14_74.pdf
  • Towards Measurement and Analysis of Large-scale Networks, AKASHI Osamu, MIZUTANI Kimihiro, FUKUDA Kensuke, URUSHIDANI Shigeo, IEICE technical report, 114, 6, 77, 82,   2014 04 17 , http://ci.nii.ac.jp/naid/110009874951
    Summary:It is important to analyze the behavior of large-scale networks. However, this analysis is difficult because we have to cope with a huge amount of and various kinds of data. A variety of applicable analysis methods make situation more complicated. We discuss a method to tackle with this problem. The method integrates network-status information in multiple layers and analyzes it from multiple viewpoints. We applies this method to analyze the behavior of SINET, which is one of the major networks in Japan. We visualize and analyze Netflow data, BGP route information, and Darknet data in an integrated fashion and discuss its effectiveness.
  • BS-6-7 Secure Virtual Network Splitting Across Multiple Infrastructure Providers, Mano Toru, Inoue Takeru, Mizutani Kimihiro, Akashi Osamu, Proceedings of the IEICE General Conference, 2014, 2, "S, 200"-"S-201",   2014 03 04 , http://ci.nii.ac.jp/naid/110009850106
  • 複数設備提供者間の情報秘匿型仮想ネットワーク分割手法, 間野暢, 井上武, 水谷后宏, 明石修, 電子情報通信学会 総合大会, BS, 6, 7,   2014 03 , http://jglobal.jst.go.jp/public/201402211272382088
  • Secure Resource Provisioning Across Multiple Domains, MANO Toru, MIZUTANI Kimihiro, AKASHI Osamu, IEICE technical report. Internet Architecture, 112, 489, 157, 162,   2013 03 14 , http://ci.nii.ac.jp/naid/110009712528
    Summary:Network resource provisioning is an important technique for infrastructure providers (infra-providers). However, to fully satisfy user requests it is probably necessary to use facilities across multiple domains, for which the conventional resource provisioning methods are unsuitable for the multiple domains because they require unre- vealed information from infraproviders. In this paper, we propose a method for resource provisioning across multiple domains that uses infraproviders ' confidential information without exposing it to other infra-providers. Evaluation results show that the computational overhead is tractable and that the average utility fee is at least on the same level as that of the conventional methods.
  • An Inplementation and Its Fundamental Evaluation of a Distributed Computing Platform Based on Structured Overlay Network, MIZUTANI Kimihiro, MANO Toru, AKASHI Osamu, FUKUDA Kensuke, IEICE technical report. Information and communication management, 112, 378, 109, 114,   2013 01 17 , http://ci.nii.ac.jp/naid/110009727350
    Summary:We propose new distributed computing platform for analyzing big data such as traffic and sensor data. The proposed platform solves the task synchronization problem and task bias problem by using ShpList tree architecture. In concrete terms, the proposed platform enables nodes to synchronize children tasks by consisting tree connections. In addition, the proposed platform realizes the task smoothing among nodes by considering the amount of nodes' task. In evaluation, we shows the concrete behevier of the proposed platform and the abalability.
  • An Inplementation and Its Fundamental Evaluation of a Distributed Computing Platform Based on Structured Overlay Network, MIZUTANI Kimihiro, MANO Toru, AKASHI Osamu, FUKUDA Kensuke, IEICE technical report. Life intelligence and office information systems, 112, 379, 109, 114,   2013 01 17 , http://ci.nii.ac.jp/naid/110009727835
    Summary:We propose new distributed computing platform for analyzing big data such as traffic and sensor data. The proposed platform solves the task synchronization problem and task bias problem by using ShpList tree architecture. In concrete terms, the proposed platform enables nodes to synchronize children tasks by consisting tree connections. In addition, the proposed platform realizes the task smoothing among nodes by considering the amount of nodes' task. In evaluation, we shows the concrete behevier of the proposed platform and the abalability.
  • B-14-16 A Study of Effective Management of Variable Bit Rate Flows, Mizutani Kimihiro, Akashi Osamu, Mano Toru, Proceedings of the IEICE General Conference, 2012, 2,   2012 03 06 , http://ci.nii.ac.jp/naid/110009463045
  • A Proposal of Scalable Publisher/Subscriber Computing Infrastructure with Structured Overlay Network, MIZUTANI Kimihiro, AKASHI Osamu, FUKUDA Kensuke, IEICE technical report. Information and communication management, 111, 382, 61, 66,   2012 01 19 , http://ci.nii.ac.jp/naid/110009480821
    Summary:We propose a new distributed computing infrastructure for managing big data such as massive traffic and sensor information effectively. Our infrastructure stores the big data onto structured overlay network and analyzes the big data through distribute cooperation. To analyze the big data through the cooperation, an administrator, who manages the big data, only have to subscribe an analysis policy for the overlay. After the subscribing, the overlay publishes the result of the analyzing to the administrator autonomously.
  • A Proposal of Scalable Publisher/Subscriber Computing Infrastructure with Structured Overlay Network, MIZUTANI Kimihiro, AKASHI Osamu, FUKUDA Kensuke, 電子情報通信学会技術研究報告. ICM, 情報通信マネジメント : IEICE technical report, 111, 382, 61, 66,   2012 01 12 , http://ci.nii.ac.jp/naid/10031099004
  • A Study of OS Kernel Functions for Estimating Available Bandwidth, KUGIMOTO Takeshi, MIYAKE Shigeki, MIZUTANI Kimihiro, MARUYAMA Mitsuru, 電子情報通信学会技術研究報告. NS, ネットワークシステム, 111, 277, 55, 60,   2011 11 03 , http://ci.nii.ac.jp/naid/10031093891
  • BS-6-31 Cooperative load balancing mechanism among data centers(BS-6. Planning, Control and Management on Networks and Services), Mizutani Kimihiro, Akashi Osamu, Terauchi Atushi, Maruyama Mitsuru, Proceedings of the Society Conference of IEICE, 2011, 2, "S, 90"-"S-91",   2011 08 30 , http://ci.nii.ac.jp/naid/110008725358
  • A Proposal of A Reinforcement Learning-Based Dynamic Session Management in Virtual Server Environments, MIZUTANI Kimihiro, KOYANAGI Izumi, TACHIBANA Takuji, IEICE technical report, 109, 378, 51, 56,   2010 01 14 , http://ci.nii.ac.jp/naid/110007999835
    Summary:In a virtualized server environment, machine resources such as CPU and memory are shared by multiple services. In such a environment, as the number of sessions for each service increases, the amount of resources that are utilized by the services increases. If thrashing occurs due to the lack of resources, the performance of the server is degraded. It is effective to estimate the amount of used resources, however, it is hard to estimate the amount of resources that are used dynamically by multiple services. In this paper, we propose a dynamic session management based on a reinforcement leaning in order to utilize the resources effectively and avoid the thrashing. In the proposed method, a learning agent estimates the amount of used resources from the response time for a service request. Then, the agent decides the acceptance or reject of all arriving session request with Q-learning. Because this method can be implemented easily in a virtualized machine, it is expected that our proposed method is used in a real environment. We evaluate the performance of our proposed method with simulation. From simulation results, we show that the proposed method can allocate the resources to multiple services effectively while avoiding the thrashing and can perform the priority control for multiple services.
  • A system for parallel estimation of multiple overlay network routing algorithms by a plug-in architecture, MIZUTANI Kimihiro, DOI Shinichi, MATSUURA Satoshi, FUJIKAWA Kazutoshi, SUNAHARA Hideki, IPSJ SIG technical reports, 2009, 8, 9, 16,   2009 01 22 , http://ci.nii.ac.jp/naid/110007137950
    Summary:As a lot of researchers study overlay network, a kind of routing algorithms are proposed. If you propose a new algorithm, you will need to prove the superiority of your algorithm. Using past estimation systems, you can estimate only one algorithm at the same time. Therefore you have to reconstruct an environment in order to estimate other algorithms. In this research, we propose a new estimation system. Our system is based on one fundamental overlay network and enables a parallel estimation of multiple routing algorithm at the same instant by a plug-in architecture. Our system offers a environment where you can estimate and compare routing algorithms on the same experimental circumstance.

Awards & Honors

  •   2017 , IEEE Global Communications Conference (GLOBECOM), Best Paper Award, A Tensor Based Deep Learning Technique for Intelligent Packet Routing
  •   2017 , 日本ソフトウェア科学会, 第21回研究論文賞, Stable Load Balancing with Overlapping ID-space Management in Range-based Structured Overlay Networks
  •   2016 , IEEE International Conference on Communications (ICC),, Best Paper Award, A Mobility-Based Mode Selection Technique for fair Spatial Dissemination of Data in Multi-Channel Device-to- Device Communication
  •   2010 , International MultiConference of Engineers and Computer Scientists(IMECS), Best Student Paper Award, Dynamic Session Management Based on Reinforcement Learning in Virtual Server Environment
  •   2010 , 情報処理学会MBL研究会, 優秀プレゼンテーション賞, An Implementation of a Framework for Integrating Churn Managements among Structured Overlay Networks

Research Grants & Projects

  • Information-technology Promotion Agency, Japan, The Mitoh Youth Project, Development of Programmable Overlay Network

Educational Activities

Teaching Experience

  • Network Technology 2, Kindai University