Search Results 1-20 of 561

  • Hybrid Coarse Grain Task Parallelization on Multi-GPU  [in Japanese]

    渡辺 智之 , 吉田 明正

    … Therefore, this paper proposes the hybrid coarse grain task parallelization scheme that utilizes GPUs for coarse grain tasks whose multicore performance is insufficient. …

    情報処理学会論文誌コンピューティングシステム(ACS) 13(3), 1-12, 2020-11-12

    IPSJ 

  • Multi-Hybrid Accelerated Simulation by GPU and FPGA on Radiative Transfer Simulation in Astrophysics

    Ryohei Kobayashi , Norihisa Fujita , Yoshiki Yamaguchi , Taisuke Boku , Kohji Yoshikawa , Makito Abe , Masayuki Umemura

    … These improvements reveal the possibility of implementing a concept to enable on-the-fly offloading computation at which CPUs/GPUs perform poorly to FPGAs while performing low-latency data movement. … These improvements reveal the possibility of implementing a concept to enable on-the-fly offloading computation at which CPUs/GPUs perform poorly to FPGAs while performing low-latency data movement. …

    情報処理学会論文誌コンピューティングシステム(ACS) 13(3), 2020-11-12

    IPSJ 

  • A New Compression Scheme of Sparse Matrix Formats for Accurate Numerical Simulation on Site Environment with GPGPU  [in Japanese]

    河村 知記 , 米田 一徳 , 岩村 尚 , 渡邉 正宏 , 井口 寧

    … Although such simulations can possibly be executed in a practical time with GPUs, some of them do not fit in GPUs due to their limited memory capacity. …

    情報処理学会論文誌数理モデル化と応用(TOM) 13(2), 93-106, 2020-08-28

    IPSJ 

  • Branch Divergence Reduction Based on Code Motion

    Junji Fukuhara , Munehiro Takimoto

    … The Single Instruction Multiple Data (SIMD) execution model on GPUs enables a program to execute efficiently. … Thisarticle should be cited as: Journal of Information Processing Vol.28(2020) (online)------------------------------The Single Instruction Multiple Data (SIMD) execution model on GPUs enables a program to execute efficiently. …

    情報処理学会論文誌プログラミング(PRO) 13(2), 2020-04-27

    IPSJ 

  • Clinical and pathological diagnosis using AI and data-driven science  [in Japanese]

    Hiraoka Shin-ichiro

    … The application of artificial intelligence (AI) technology in healthcare is spreading rapidly thanks to the construction of a medical "big data" database as well as the development of graphics processing units (GPUs) that can process data at high speed. …

    Journal of Japanese Society of Oral Oncology 32(4), 159-170, 2020

    J-STAGE 

  • Multi-Hybrid Accelerated Simulation by GPU and FPGA on Radiative Transfer Simulation in Astrophysics

    Kobayashi Ryohei , Fujita Norihisa , Yamaguchi Yoshiki , Boku Taisuke , Yoshikawa Kohji , Abe Makito , Umemura Masayuki

    … These improvements reveal the possibility of implementing a concept to enable on-the-fly offloading computation at which CPUs/GPUs perform poorly to FPGAs while performing low-latency data movement. …

    Journal of Information Processing 28(0), 1073-1089, 2020

    J-STAGE 

  • A Data-Centric Directive-Based Framework to Accelerate Out-of-Core Stencil Computation on a GPU

    SHEN Jingcheng , INO Fumihiko , FARRÉS Albert , HANZICH Mauricio

    … <p>Graphics processing units (GPUs) are highly efficient architectures for parallel stencil code; …

    IEICE Transactions on Information and Systems E103.D(12), 2421-2434, 2020

    J-STAGE 

  • FiC-RNN: A Multi-FPGA Acceleration Framework for Deep Recurrent Neural Networks

    SUN Yuxi , AMANO Hideharu

    … This inefficiency manifests itself in the proportional increase in the latency of RNN inference with respect to the number of layers of deep RNNs on CPUs and GPUs. …

    IEICE Transactions on Information and Systems E103.D(12), 2457-2462, 2020

    J-STAGE 

  • Block Randomized Singular Value Decomposition on GPUs

    LU Yuechao , MATSUSHITA Yasuyuki , INO Fumihiko

    … For processing large-scale data, computing systems with accelerators like GPUs have become the mainstream approach. …

    IEICE Transactions on Information and Systems E103.D(9), 1949-1959, 2020

    J-STAGE 

  • Tensegrity representation of microtubule objects using unified particle objects and springs

    Pramudwiatmoko Arif , Gutmann Gregory , Ueno Yutaka , Kakugo Akira , Yamamura Masayuki , Konagaya Akihiko

    … A particle simulation system that utilizes multiple GPUs resources is used to fulfill haptic VR requirements. …

    Chem-Bio Informatics Journal 20(0), 19-43, 2020

    J-STAGE 

  • Fast Euclidean Cluster Extraction Using GPUs

    Nguyen Anh , Cano Abraham Monrroy , Edahiro Masato , Kato Shinpei

    <p>Clustering is the task of dividing an input dataset into groups of objects based on their similarity. This process is frequently required in many applications. However, it is computationally …

    Journal of Robotics and Mechatronics 32(3), 548-560, 2020

    J-STAGE 

  • TensorShader : Deep Learning Framework for High-Dimensional Neural Networks  [in Japanese]

    YOSHIMURA Takuma

    … On the other hand, there are still few deep learning frameworks that can handle high-dimensional neural networks on GPUs, which hinders experiments. …

    Proceedings of the Annual Conference of JSAI JSAI2020(0), 1J5GS201-1J5GS201, 2020

    J-STAGE 

  • Automatic Detection of Marine Plastic by Composite Remote Sensing with Deep Learning  [in Japanese]

    SONODA Jun , KIMOTO Tomoyuki , KANAZAWA Yasushi

    … We have generated the GPR images for training using a fast finite-difference time-domain (FDTD) simulation with graphics processing units (GPUs). …

    Proceedings of the Annual Conference of JSAI JSAI2020(0), 1D4GS1303-1D4GS1303, 2020

    J-STAGE 

  • Locally mesh-refined lattice Boltzmann method for fuel debris air cooling analysis on GPU supercomputer

    ONODERA Naoyuki , IDOMURA Yasuhiro , UESAWA Shinichiro , YAMASHITA Susumu , YOSHIDA Hiroyuki

    … It is also shown that the AMR based CityLBM code on 4 NVIDIA TESLA V100 GPUs gives 6.7x speedup of the time to solution compared with the JUPITER code on 36 Intel Xeon E5-2680v3 CPUs. …

    Mechanical Engineering Journal 7(3), 19-00531-19-00531, 2020

    J-STAGE 

  • Branch Divergence Reduction Based on Code Motion

    Fukuhara Junji , Takimoto Munehiro

    … <p>The Single Instruction Multiple Data (SIMD) execution model on GPUs enables a program to execute efficiently. …

    Journal of Information Processing 28(0), 302-309, 2020

    J-STAGE 

  • A Highly Configurable 7.62GOP/s Hardware Implementation for LSTM

    FAN Yibo , HUANG Leilei , CHEN Kewei , ZENG Xiaoyang

    … Long Short-Term Memory (LSTM), a popular type of recurrent neural networks (RNNs), has been widely implemented on CPUs and GPUs. …

    IEICE Transactions on Electronics E103.C(5), 263-273, 2020

    J-STAGE 

  • Construction of environment for education and research using machine learning(Spesial Issue)  [in Japanese]

    佐藤 進也

    … Then, we describe the overall architecture of the environment constructed by combining the selected elemental technologies, which include Python, Jupyter, Docker, and Linux servers with GPUs. …

    日本工業大学研究報告 = Report of researches, Nippon Institute of Technology 49(3), 34-36, 2019-12

    IR 

  • Detecting System Failures with GPUs and LLVM

    Ozaki Yuichi , Kanamoto Sousuke , Yamamoto Hiroaki , Kourai Kenichi

    … This paper proposes GPUSentinel for more reliable white-box monitoring using general-purpose GPUs. … Since GPUs are isolated from the target system, system monitors are not easily affected by system failures. …

    APSys '19: Proceedings of the 10th ACM SIGOPS Asia-Pacific Workshop on Systems, 47-53, 2019-08

    IR 

  • Formal Approach to Editing a Tensorflow Computational Graph for Large Model Support

    Tung D. Le , Haruki Imai , Yasushi Negishi , Kiyokuni Kawachiya

    … While accelerators such as GPUs are suitable for training neural networks, they have limited memory. … Moreover, modern IBM machines for AI are integrated with NVLinks that provide very fast connection between CPUs and GPUs. … While accelerators such as GPUs are suitable for training neural networks, they have limited memory. … Moreover, modern IBM machines for AI are integrated with NVLinks that provide very fast connection between CPUs and GPUs. …

    情報処理学会論文誌プログラミング(PRO) 12(2), 17-17, 2019-05-21

    IPSJ 

  • Multi-GPUs parallelization for phase-field simulations of dendrite growth applying an adaptive mesh refinement method  [in Japanese]

    坂根 慎治 , 高木 知弘 , 大野 宗一 , 澁田 靖 , 青木 尊之

    計算工学講演会論文集 Proceedings of the Conference on Computational Engineering and Science 24, 4p, 2019-05

Page Top