博 士 学 位 論 文 Doctoral Dissertation
内容の要旨 及び 審査結果の要旨
Dissertation Abstract and
Summary of the Dissertation Review Result
第32号
The Thirty-Second Issue
2018年9月 September, 2018 The University of Aizu
はしがき
博士の学位を授与したので、学位規則（昭和２８年４月１日文部省令第９号）第８条の規定 に基づき、その論文の内容の要旨及び論文審査の結果の要旨をここに公表する。
学位記番号に付した｢甲｣は学位規則第４条第１項（いわゆる課程博士）によるものであるこ とを示す。
Preface
On granting the Doctoral Degree to the individuals mentioned below, abstracts of their theses and the theses review results are herewith publicly announced, in according to the provisions provided for in Article 8 of the Ruling of Degrees (Ministry Of Education Ordinance No.9, enacted on April 1, 1953)
The Chinese character, “甲”, at the beginning of the diploma number represents that an individual has been granted the degree in accordance with the provisions provided for in Paragraph 4-1 of the Ruling Of Degrees (what is called “Katei Hakase,” or the Doctoral Degree granted by the University at which the grantee was enrolled.).
- 1 -
目 次 Contents
掲載順
Order
学位記番号 Diploma No.
学位 Degree
氏名 Name
論文題目 Dissertation Title
頁 Page
1
甲CI博 第64号
博士（コンピュー タ理工学）
The Degree of Doctor of Science and
Engineering
KONO, Fumiya 河野 郁也
Many-core CPU, GPU, FPGAによる高性 能かつポータブルな津波シミュレーション High Performance Portable Tsunami Simulations on Many-core CPU, GPU, and FPGA
2
2
甲CI博 第65号
博士（コンピュー タ理工学）
The Degree of Doctor of Science and
Engineering
RYSKELDIEV, Bektur
共同作業ソフトウェアにおける周囲のパノ ラマ閲覧，状況認識，および豊かなユーザ ー体験を提供するためのリアルタイムメデ ィアストリームの合成
Compositing real-time media streams for groupware panoramic browsing, situation awareness, and enriched user experience
5
- 2 - Name
氏名
KONO, Fumiya 河野 郁也 The relevant degree
学位の種類
Doctoral degree (in Computer Science and Engineering) 博士（コンピュータ理工学）
Number of the diploma of the Doctoral Degree 学位記番号
甲CI博第64号
The Date of Conferment 学位授与日
September 19, 2018 2018年9月19日 Requirements for Degree Conferment
学位授与の要件
Please refer to the article five of “University Regulation on University Degrees”
会津大学学位規程 第5条該当 Dissertation Title
論文題目
High Performance Portable Tsunami Simulations on Many-core CPU, GPU, and FPGA
Many-core CPU, GPU, FPGAによる高性能かつポータブルな 津波シミュレーション
Dissertation Review Committee Members 論文審査委員
The University of Aizu, Senior Associate Prof. NAKASATO, N.
(Chief Referee)
The University of Aizu, Prof. VAZHENIN, A. P.
The University of Aizu, Senior Associate Prof. ASAI, N.
The University of Aizu, Associate Prof. MATSUMOTO, K.
会津大学上級准教授 中里 直人（主査）
会津大学教授 アレクサンンダー ヴァジェニン 会津大学上級准教授 浅井 信吉
会津大学准教授 松本 和也
- 3 -
Abstract
Tsunami generated by a submarine earthquake sometimes causes serious damage in a coastal area. To reduce negative effects of tsunami, effective evacuation and disaster prevention are getting interested.
We can contribute to that by forecasting arrival time and height of tsunami with computer simulations.
However, tsunami simulations always require massive data processing. The shallow water equations for tsunami modeling require wave height, wave speed and depth of the sea for each computation grid.
The total number of computation grid also becomes over several millions. Though a sequential computation with a single-core CPU can complete tsunami simulation, technologies to complete the simulation as fast as possible are desired to reduce the damage of tsunami.
In modern computer systems, various architectures for parallel computations are presented. Modern CPUs are designed as multicore systems.
GPUs (Graphic Processing Units) were initially introduced to accelerate image processing. Since GPUs are also expected high performance for parallel computations, they are now applied to accelerate the general computations (GPGPU). FPGA is also attractive in regard to the compatibility of high performance computation and low power consumption. As such modern architecture appears, the parallel computing technologies such as OpenMP, OpenACC, CUDA, and OpenCL are also presented.
In this dissertation, we developed various kinds of parallel codes which aim to accelerate the MOST algorithm for tsunami modeling. We conducted performance benchmarking of our parallel codes on various modern architectures such as Intel Xeon, Intel Xeon Phi, NVIDIA Tesla GPU, AMD FirePro GPU, AMD Radeon GPU, and Arria 10 FPGA. We evaluated the performance of each computation and investigated optimal implementation for the MOST algorithm.
Currently, the best result is achieved by using OpenCL kernel with no optimization on AMD Radeon R9 280X GPU whose performance is 185GFlops. The computation time is 2.41 seconds for 300 time-steps which corresponds to 5 minutes in real-time. Therefore, our computation by using OpenCL and Radeon GPU is applicable to the real tsunami prediction system.
The implementation of FPGA design presented in this dissertation is based on the OpenCL kernel programming. The technology which generates FPGA designs from OpenCL kernels known as High-level synthesis (HLS) is recently getting practical. We here evaluated the performance of FPGA designs generated by a compiler supported by Intel. To achieve better performance on FPGA, we optimized our GPU kernel codes for FPGA by implementing loop-unrolling so that the compiler can exploit shift registers for the computation.
The performance of our FPGA design is improved by an implementation to compute multiple grid points on one pipeline. Furthermore, the methodology of HLS is even getting sophisticated in these years. We compared FPGA designs generated by two compilers in regard to performance, resource utilization, and efficiency of floating-point operations. The performance of a design by a new compiler reaches to 153Gflops which is more than twice as much as a design by an old one.
Finally, we discussed the applicability of our parallel implementations to the real-time tsunami
- 4 -
simulation based on phase velocity of the wave which derives from shallow water equations. We here used the result of the OpenCL kernel on Radeon GPU which achieved the highest performance of all combinations. We first showed scalability of our computation and calculated the computation time for updating one computation grid. Afterwards, we tested our implemented OpenCL kernel under some initial conditions referring to past earthquakes and tsunami; the earthquake near the west coastal region of America in 2005, and the 2011 Tohoku earthquake and tsunami in Japan. Estimated computation time for each situation is enough fast compared to actual arrival time of tsunami. In regard to the computation time required for numerical simulations, we can conclude that performance of our implementation is sufficient for real-time tsunami simulation.
Summary of the Dissertation Review Result
The dissertation presents the acceleration of a tsunami modeling code MOST on various parallel computer architectures as Many-core, Graphic Processing Unit (GPU), and Field Programmable Gate Array (FPGA) systems through programming framework such as OpenMP, OpenACC, CUDA and OpenCL. For GPU systems, the author compared the performance of CUDA, OpenACC and OpenCL implementations and analyzed the performance bottleneck on GPUs. Furthermore, the author implemented the MOST on FPGA through the OpenCL based high-level synthesis (HLS) tool for the first time. On the evaluated recent FPGA, the author found that a pipelined implementation using a buffer structure called the shift-register is most effective. Finally, the author evaluated the practicability of the real-time simulation of MOST on those parallel computer architectures.
Accordingly, the MOST implementation in OpenCL framework enabled high performance and portable tsunami simulations on computer systems attached GPU / FPGAs in real-time.
In the final review, the candidate presented his work in 70 minutes followed by 50 minutes questions and discussion. The committee have reviewed the submitted dissertation and the response to questions raised after the preliminary review and satisfied the answers. All member of the committee agreed to confirm the significance of the dissertation for a PhD degree.
- 5 - Name
氏名
RYSKELDIEV, Bektur
（りすけりでぃえふ べくとぅる）
The relevant degree 学位の種類
Doctoral degree (in Computer Science and Engineering) 博士（コンピュータ理工学）
Number of the diploma of the Doctoral Degree 学位記番号
甲CI博第65号
The Date of Conferment 学位授与日
September 19, 2018 2018年9月19日 Requirements for Degree Conferment
学位授与の要件
Please refer to the article five of “University Regulation on University Degrees”
会津大学学位規程 第5条該当 Dissertation Title
論文題目
Compositing real-time media streams for groupware
panoramic browsing, situation awareness, and enriched user experience
共同作業ソフトウェアにおける周囲のパノラマ閲覧，状況 認識，および豊かなユーザー体験を提供するためのリアル タイムメディアストリームの合成
Dissertation Review Committee Members 論文審査委員
The University of Aizu, Prof. COHEN, M.(Chief Referee)
The University of Aizu, Associate Prof. VILLEGAS, J.
The University of Aizu, Associate Prof. MOZGOVOY, M.
The University of Aizu, Prof. LUBASHEVSKIY, I.
会津大学教授 マイケル コーエン（主査）
会津大学准教授 ヴィジェガス ジュリアン 会津大学准教授 マキシム モズゴボイ 会津大学教授 イゴール ルバシェフスキー
- 6 -
Abstract
According to the Cisco report published in June 2017 [1], by year 2021 82% of the world’s IP traffic will be taken over by video streaming services. Within that segment, live video streaming (often referred to as “livestreaming”) would represent 13% of the world’s video traffic, with mobile users being the fastest growing group of consumers. It is evident that video streaming is growing in popularity, which perhaps can be explained by its relative ease of use: often in video streaming applications users are a single “tap” away from experiencing the media content. Such availability makes livestreaming, or more specifically, mobile livestreaming, ubiquitous in cases where a quick sharing of a remote situation with multiple users in real time is needed, including social and collaborative applications such as Facebook Live, Periscope, Skype, and FaceTime.
However, such convenience comes with a cost: usually a streaming session consists of a lowresolution single-viewpoint video stream, which provides limited information about a remote situation. In scenarios where spatial context is important, for instance when a streaming user constantly changes locations and interacts with different environments, such limitations can negatively affect user experience: it is hard for viewers to understand where a streaming user is located, and for streamers it is equally hard to react when viewers refer to a certain part of the environment surrounding the streamer.
Therefore, this dissertation investigates and proposes a new approach to how mobile live media streams are presented and interacted with in social and collaborative applications. The main research problem can be formulated in the following question: “Given the spatial data inferred from mobile devices, how can multiuser live video streaming sessions be improved?” In this case the definition of
“improvement” is dispersed and discussed within the scope of the following contributions.
In the first contribution, this thesis approaches the livestream composition problem from the perspectives of mixed reality displays and social livestreaming systems. It proposes new and updates existing taxonomies, upon which the related publications and projects were categorized and qualitatively compared.
Based on the observed literature, in the second contribution this thesis outlines and proposes a new spatial livestream composition (SLC) method, which organizes spatial and media information into two categories: spatial background, which provides a generalized spatial context, and spatialized live media streams which are embedded within the spatial background, creating a composited mixed reality space in which users can experience multiple live media streams and interact with each other in real time.
The proposed SLC method was implemented and evaluated in several proof-of-concept applications.
In the third contribution, this thesis discusses an implementation of SLC method in a collaborative mixed reality application, which was tested in a user study that detected a statistically significant decrease in mental workload and increase in spatial and situational awareness among viewers in comparison with a regular video streaming application. Based on these results, in the fourth contribution this dissertation applies the SLC method to social interactive applications, and investigates whether the proposed method also affects user engagement in a user study that compares
- 7 -
the developed application with Periscope (a popular social livestreaming platform). Although the user comments favored the social livestreaming application with SLC method, the statistical results were inconclusive, and therefore a set of recommendations for similar studies was formed.
In the final contribution, this dissertation discusses how SLC method can be combined with other mixed reality applications by proposing a decentralized “metaverse,” a blockchain-based method for organization and sharing of virtual spaces for mixed reality systems. It discusses how this approach can be integrated in both SLC and non-SLC based systems, and their possible future applications.
Summary of the Dissertation Review Result
This dissertation investigates and proposes a new approach to how mobile live media streams are presented and interacted with in social and collaborative applications. The main research problem can be formulated in the following question: "Given the spatial data inferred from mobile devices, how can multiuser live video streaming sessions be improved?" In this case the definition of
"improvement" is discussed within the scope of the following contributions.
In the first contribution, this dissertation approaches the livestream composition problem from the perspectives of mixed reality displays and social livestreaming systems. It proposes new and updates existing taxonomies, upon which the related publications and projects were categorized and qualitatively compared.
Based on the observed literature, in the second contribution this dissertation outlines and proposes a new spatial livestream composition (SLC) method, which organizes spatial and media information into two categories: spatial background, which provides a generalized spatial context, and spatialized live media streams which are embedded within the spatial background, creating a composited mixed reality space in which users can experience multiple live media streams and interact with each other in real time.
The proposed SLC method was implemented and evaluated in several proof-of-concept applications.
In the third contribution, this dissertation discusses an implementation of SLC method in a collaborative mixed reality application, which was tested in a user study that detected a statistically significant decrease in mental workload and increase in spatial and situational awareness among viewers in comparison with a regular video streaming application. Based on these results, in the fourth contribution this dissertation applies the SLC method to social interactive applications, and investigates whether the proposed method also affects user engagement in a user study that compares the developed application with Periscope (a popular social livestreaming platform). Although the user comments favored the social livestreaming application with SLC method, the statistical results were inconclusive, and therefore a set of recommendations for similar studies was formed.
In the final contribution, this dissertation discusses how SLC method can be combined with other mixed reality applications by proposing a decentralized "metaverse," a blockchain-based method for organization and sharing of virtual spaces for mixed reality systems. It discusses how this approach can be integrated in both SLC and non-SLC based systems, and their possible future applications.
- 8 -
In the final review, the candidate presented a 2-hour summary interrupted or followed by Q&A.
The candidate has an appropriate level of understanding of his area of research and scientific exploration. He displays a solid knowledge of his field and its related areas, an understanding of "the big picture," and has many ideas regarding interesting research vectors for the future.
All the issues raised at the preliminary review were addressed.
The committee reviewed and unanimously approved the submitted dissertation.
- 9 -
博 士 学 位 論 文 Doctoral Dissertation
内容の要旨 及び 審査結果の要旨 Dissertation Abstract
and
Summary of the Dissertation Review Result
第32号
The Thirty-Second Issue
2018年9月 September, 2018
発行 会津大学
〒965-8580 福島県会津若松市一箕町鶴賀 TEL: 0242-37-2600
FAX: 0242-37-2526 THE UNIVERSITY OF AIZU Tsuruga, Ikki-machi Aizu-Wakamatsu City
Fukushima, 965-8580 Japan
