2018

2018 Conference on Advanced Topics and Auto Tuning in

High-Performance Scientific Computing

March 26-27, 2018

Lecture Room, 1F Management Building

National Cheng Kung University, Tainan, Taiwan

Plenary Speaker

  • Chien-Min Kao (University of Chicago)

  • Jose E. Roman (Universitat Politècnica de València)

Invited Speakers

  • Akihiro Fujii (Kogakuin University)

  • Takeshi Fukaya (Hokkaido University)

  • Chieh-Sen Huang (National Sun Yat-sen University)

  • Feng-Nan Hwang (National Central University)

  • Akihiro Ida (The University of Tokyo)

  • Toshiyuki Imamura (RIKEN AICS)

  • Takahiro Katagiri (Nagoya University)

  • Gwo Giun (Chris) Lee (National Cheng Kung University)

  • Ching-Sung Liu (National University of Kaohsiung)

  • Darsen Lu (National Cheng Kung University)

  • Kouki Masui (Nagoya University)

  • Toru Nagai (Nagoya University)

  • Kengo Nakajima (The University of Tokyo)

  • Naoya Nomura (The University of Tokyo)

  • Takeshi Ogita (Tokyo Woman's Christian University)

  • Kenji Ono (Kyusyu University)

  • Katsuhisa Ozaki (Shibaura Institute of Technology)

  • Matthew Smith (National Cheng Kung University)

  • Reiji Suda (The University of Tokyo)

  • Tomohiro Suzuki (Yamanashi University)

  • Daisuke Takahashi (Tsukuba University)

  • Hiroyuki Takizawa (Tohoku University)

  • Teruo Tanaka (Kogakuin University)

  • Chun-Hao Teng (National Chung Hsing University)

  • Takeshi Terao (Shibaura Institute of Technology)

  • Yun-Che Wang (National Cheng Kung University)

  • Chin-Tien Wu (National Chiao Tung University)

Program Committee

Local Committee

Aims and Scope

The Conference on Advanced Topics and Auto Tuning in High-Performance Scientific Computing focuses on the scientific impacts due to the latest computer architectures and the approaches to achieve high-performance computing on these leading-edge computers. Advances in many-core architectures and high-end computers have unveiled their significances in the scientific discoveries and engineering achievements. The complexity of these newly developed computers, however, also leads to contemporary challenges to achieve the best efficiency of the highly promising computational capabilities. The conference encourages interdisciplinary communications between researchers from applied mathematics, statistics, computer science, physical sciences, engineering and industry to prompt innovations and breakthroughs in this exciting field. The main themes include, but not limited to, simulations, numerical methods, applications, hardware, and particularly software and algorithm auto-tuning via statistical methods.

Contact Person

Prof. Ray-Bin Chen (raybingc@gmail.com)

Sponsors

02 Program

Download the Program Booklet PDF

Program

March 26, 2018 (Monday)

09:00 - 09:10 Opening

09:10 - 10:30 Session 1: HPSC Algorithms and Its Scientific Applications: Part I (Chair: Takahiro Katagiri)

  • Gwo Giun (Chris) Lee (National Cheng Kung University)

  • Algorithm/Architecture Co-design: From System-on-Chip to Internet-of-Things and Cloud/Edge

  • Hiroyuki Takizawa (Tohoku University)

  • Automatic Parameter Tuning of Application-level Incremental Checkpointing

  • Reiji Suda and Yushi Qiu (The University of Tokyo)

  • Online Autotuning of Parameters of GA-BiCG

  • Kenji Ono (Kyusyu University)

  • Dynamic Selection of the Number of Padding to Improve the Performance Caused by Slushing

Group Photo and Coffee Break

11:00 - 12:00 Session 2: Novel HPSC methods on Parallel Computing (Chair: Min-Hung Chen)

  • Matthew Smith (National Cheng Kung University)

  • Dual Layer Parallel Computation of Compressible Shocked Flows using the AMD Ryzen Threadripper CPU

  • Takeshi Fukaya (Hokkaido University)

  • Performance Evaluation of Time-Space Tiling with Tile-Level Parallelism for Iterative Stencil Computations

  • Chieh-Sen Huang (National Sun Yat-Sen University)

  • On the Parallel Algorithms of the Near-Optimal Solution of Vehicle Routing Problems

Lunch Break

13:30-14:30 Plenary Talk I (Chair: Reiji Suda)

Jose E. Roman (Universitat Politècnica de València)

Current Trends in the Parallel Solution of Large-Scale Eigenvalue Problems

14:30 - 15:30 Session 3: Matrix Decomposition and Tensor Analysis (Chair: Yueh-Cheng Kuo)

  • Ching-Sung Liu (National University of Kaohsiung)

  • Newton-Noda Iteration for Finding the Perron Pair of an Irreducible Nonnegative Tensor

  • Toshiyuki Imamura (RIKEN AICS)

  • Review of Tensor-Decomposition Toolkit on K Computer

  • Tomohiro Suzuki (Yamanashi University)

  • Implementation of Tile Matrix Factorization and Its Performance Model

Coffee Break

16:00 - 17:20 Session 4: Advanced Methods for Differential Equations and Its Applications (Chair: Takeshi Fukaya)

  • Feng-Nan Hwang (National Central University)

  • A Multiscale Finite Element Method with Locally Adaptive Bubble Function Enrichment

  • Toru Nagai (Nagoya University)

  • An Introduction to a New Method for Solving Wave Equation and Numerical Demonstrations

  • Chun-Hao Teng (National Chung Hsing University)

  • Numerical Modeling of Poisson-Nernst-Plank Type Models by Spectral Element Methods

  • Chin-Tien Wu (National Chiao Tung University)

  • A Skew Freeform Reflector Design Method for Two Dimensional Illumination

17:30 - 20:00 Banquet

March 27, 2017 (Tuesday)

9:10 - 10:30 Session 5: HPSC Algorithms and Its Scientific Applications: Part II (Chair: Chern-Shuh Wang)

  • Darsen Lu (National Cheng Kung University)

  • Enabling Electronic Circuit Simulation for New Semiconductor Technologies -- Using the Double-Gate MOS Transistor as an Example

  • Yun-Che Wang (National Cheng Kung University)

  • Study of Mechanical and Thermal Properties of Calcium-Silicate-Hydrate via Molecular Dynamics Simulation

  • Takeshi Terao (Shibaura Institute of Technology)

  • Verification of Positive Definiteness of Symmetric Sparse Matrices

  • Takeshi Ogita (Tokyo Woman's Christian University)

  • Development of Verified Numerical Computations in High-Performance Computing Environments

Coffee Break

11:00 - 12:00 Session 6: Efficient HPSC techniques on Automatic Tuning (Chair: Yin-Liang Huang)

  • Kengo Nakajima (The University of Tokyo)

  • Parallel Multigrid Method with Automatic Tuning for Manycore Clusters

  • Takahiro Katagiri (Nagoya University)

  • Auto-Tuning of Preconditioner Selection for Sparse Iterative Solvers --- Adaptation of Deep Learning and its Limitations

  • Teruo Tanaka (Kogakuin University)

  • Low Cost Multiple One-dimensional d-Spline Parameter Search for Multiple Performance Parameter Estimation

Lunch Break

13:30 - 14:30 Plenary Talk II (Chair: Weichung Wang)

  • Chien-Min Kao (University of Chicago)

    • Selected Topics of Optimization, Computing and AI in Biomedical Imaging

14:30 - 15:30 Session 7: Advanced Techniques for Solving Sparse and Large-scale Linear systems (Chair: Chien-Hong Cho)

  • Akihiro Fujii (Kogakuin University)

  • Acceleration of Sparse Linear Solvers Using RDMA Communication on PRIMEHPC FX Supercomputer

  • Katsuhisa Ozaki (Shibaura Institute of Technology)

  • Verified Numerical Solutions of Very-Large-Scale Linear Systems

  • Naoya Nomura (The University of Tokyo)

  • The Analysis for Robustness of SA-AMG Method by Using Extraction method of Near-Kernel Vectors

Coffee Break

16:00 - 17:00 Session 8: Advanced HPSC Techniques and Its applications (Chair: Ray-Bing Chen)

  • Daisuke Takahashi (Tsukuba University)

  • Implementation of Parallel 1-D Real FFT on Intel Xeon Phi Processors Authors

  • Kouki Masui (Nagoya University)

  • Performance Evaluation of Some Iterative Methods for Electromagnetic Field Analysis Using Mixed Precision Calculation

  • Akihiro Ida (The University of Tokyo)

  • Lattice H-matrices : A New Efficient Variant on Distributed Memory Systems

17:00 - 17:20 Closing

03 Registration

You are welcome to join the conference. There is no need to register in advance.

04 Information

Hotel and tour information.

05 Poster