| 09:30-17:00 (Lobby/Auditorium, Kobe U.) |
(Registration) |
| Workshop: The Seventh International Workshop on Automatic Performance Tuning (iWAPT 2012) | |
| 09:50-18:40 (Auditorium, Kobe U.) |
The Seventh International Workshop on Automatic Performance Tuning (iWAPT 2012) [Program] |
| Tutorial: High-Performance Numerical Tools for the Development and Scalability of High-End Computer Applications | |
| 09:50-18:20 (3F Seminar Room, Kobe U.) |
High-Performance Numerical Tools for the Development and Scalability of High-End Computer Applications |
| 08:30-17:00 (Lobby/Auditorium, Kobe U.) |
(Registration) |
| 11:30-17:00 (Lobby/Auditorium, Kobe U.) |
(Industry Exhibitions) |
| Opening Session (Auditorium, Kobe U.), Chair: Kengo Nakajima (The University of Tokyo, Japan) | |
| 09:15-09:25 (Auditorium, Kobe U.) |
Opening Remarks Michel Dayde (ENSEEIHT, France)(Chair, Steering Committee of VECPAR 2012) |
| 09:25-09:45 | Welcome Address: Overview of K computer and RIKEN AICS Kimihiko Hirao (Director, RIKEN AICS, Japan)(Co-Chair, Organizing Committee of VECPAR 2012) |
| 09:45-10:00 | Announcements |
| Invited Talk-1 (Auditorium, Kobe U.), Chair: Mitsuhisa Sato (University of Tsukuba, Japan) | |
| 10:00-10:45 (Auditorium, Kobe U.) |
Barriers to Exascale Computing Horst Simon (Lawrence Berkeley National Laboratory, USA) |
| Invited Talk-2 (Auditorium, Kobe U.), Chair: Kimihiko Hirao (RIKEN AICS, Japan) | |
| 10:45-11:30 (Auditorium, Kobe U.) |
Materials Design through Computics: Nanostructures of Silicon and Carbon Atsushi Oshiyama (The University of Tokyo, Japan) |
| 11:30-12:45 | (Lunch Break) |
| Session 1-A: GPU Computing (I) (Auditorium, Kobe U.), Chair: Satoshi Matsuoka (Tokyo Institute of Technology, Japan) | |
| 12:45-13:15 (Auditorium, Kobe U.) |
Programming the LU Factorization for a Multicore
System with Accelerators Jakub Kurzak (University of Tennessee, USA), Piotr Luszczek (University of Tennessee, USA), Mathieu Faverge (University of Tennessee, USA) and Jack Dongarra (University of Tennessee/Oak Ridge National Laboratory/University of Manchester, USA) |
| 13:20-13:50 | Efficient Two-Level Preconditioned Conjugate
Gradient Method on the GPU Rohit Gupta (Delft University of Technology, Netherlands), Martin B. Van Gijzen (Delft University of Technology, Netherlands) and Cornelis Vuik (Delft University of Technology, Netherlands) |
| 13:55-14:25 | Parallel Algorithm for the QR Decomposition with
Pivoting in Multicore and GPU Processors Andres Tomas (University of California, Davis, USA), Zhaojun Bai (University of California, Davis, USA) and Vicente Hernandez (Universitat Politecnica de Valencia, Spain) |
| Session 1-B: Applications (I) (Seminar Room, RIKEN AICS), Chair: Takashi Furumura (The University of Tokyo, Japan) | |
| 12:45-13:15 (Seminar Room, RIKEN AICS) |
Numerical Simulation of Long-term Fate of CO2
Stored in Deep Reservoir Rocks on Massively Parallel Vector
Supercomputer Hajime Yamamoto (Taisei Corporation, Japan), Shinichi Nanai (Taisei Corporation, Jaoab), Keni Zhang (Beijing Normal University, China), Pascal Audigane (BRGM, France), Christophe Chiaberge (BRGM, France), Ryusei Ogata (NEC Corporation, Japan), Noriaki Nishikawa (JAMSTEC, Japan), Yuichi Hirokawa (JAMSTEC, Japan), Satoru Shingu (JAMSTEC, Japan) and Kengo Nakajima (The University of Tokyo) |
| 13:20-13:50 | High Performance Simulation of Complicated Fluid
Flow in 3D Fractured Porous Media Using LBM Jinfang Gao (The University of Queensland, Australia) and Huilin Xing (The University of Queensland, Australia) |
| 13:55-14:25 | Parallel scalability enhancements of seismic
response and evacuation simulations of IES Maddegedara Lalith (The University of Tokyo, Japan), Hori Muneo (The University of Tokyo, Japan) and Ichimura Tsuyoshi (The University of Tokyo, Japan) |
| 14:25-14:50 (Lounge/Auditorium, Kobe U.) |
(Coffee Break) |
| Invited Talk-3 (Auditorium, Kobe U.), Chair: Takeshi Iwashita (Kyoto University, Japan) | |
| 14:50-15:35 (Auditorium, Kobe U.) |
Peta scale FDM Simulation of Strong Ground Motion and Tsunami: Towards Disaster Prediction and Mitigation Takashi Furumura (The University of Tokyo, Japan) |
| Session 2-A (Auditorium, Kobe U.): GPU Computing (II), Chair: Jakub Kurzak (University of Tennessee, USA) | |
| 15:40-16:10 (Auditorium, Kobe U.) |
A High Performance SYMV Kernel on a Fermi-core GPU Toshiyuki Imamura (The University of Electro-Communications, Japan), Susumu Yamada (Japan Atomic Energy Agency, Japan) and Masahiko Machida (Japan Atomic Energy Agency, Japan) |
| 15:15-16:45 | Optimizing Memory-Bound SYMV Kernel on GPU Hardware Accelerators Ahmad Abdelfattah (KAUST, Saudi Arabia), David Keyes (KAUST, Saudi Arabia), Jack Dongarra (University of Tennessee, USA) and Hatem Ltaief (KAUST, Saudi Arabia) |
| Session 2-B (Seminar Room, RIKEN AICS): Applications (II), Chair: Hideyuki Usui (Kobe University, Japan) | |
| 15:40-16:10 (Seminar Room, RIKEN AICS) |
Martin Larsson (ONERA, France) and Olivier Marquet (ONERA, France) |
| 16:15-16:45 | QMC=Chem: a quantum Monte Carlo program for
large-scale simulations in chemistry at the petascale level and beyond Anthony Scemama (Laboratoire de Chimie et Physique Quantiques CNRS/IRSAMC, France), Michel Caffarel (Laboratoire de Chimie et Physique Quantiques CNRS/IRSAMC, France), Emmanuel Oseret (GENCI-CEA-INTEL-UVSQ, France) and William Jalby (GENCI-CEA-INTEL-UVSQ, France) |
| Tour to K computer | |
| 16:55-17:40 (Seminar Room, RIKEN AICS) |
The tour starts from the Seminar Room, RIKEN AICS. Please move to RIKEN AICS building. |
| 08:30-17:00 (Lobby/Auditorium, Kobe U.) |
(Registration) |
| 09:00-16:30 (Lobby/Auditorium, Kobe U.) |
(Industry Exhibitions) |
| Invited Talk-4 (Auditorium, Kobe U.), Chair: Horst Simon (Lawrence Berkeley National Laboratory, USA) | |
| 09:00-09:45 (Auditorium, Kobe U.) |
Grand Challenge in Life Science on K computer Ryutaro Himeno (RIKEN, Japan) |
| 09:45-10:10 (Lounge/Auditorium, Kobe U.) |
(Coffee Break) |
| Session 3-A (Auditorium, Kobe U.): Finite Element Method from Various Viewpoints, Chair: Leroy A. Drummond (Lawrence Berkeley National Laboratory, USA) | |
| 10:10-10:40 (Auditorium, Kobe U.) |
A mass conservation algorithm for adaptive unrefinement in finite element methods Jing-Ru C. Cheng (U.S. Army Engineer Research and Development Center, USA), Hung V. Nguyen (U.S. Army Engineer Research and Development Center, USA), Charlie R. Berger (U.S. Army Engineer Research and Development Center, USA) and Gaurav Savant (U.S. Army Engineer Research and Development Center, USA) |
| 10:45-11:15 | Optimizing Sparse Matrix Assembly in Finite
Element Solvers with One-Sided Communication Niclas Jansson (KTH Royal Institute of Technology, Sweden) |
| 11:20-11:50 | Implementation and Evaluation of 3D Finite
Element Method Application for CUDA Satoshi Ohshima (The University of Tokyo, Japan), Masae Hayashi (The University of Tokyo, Japan), Takahiro Katagiri (The University of Tokyo, Japan) and Kengo Nakajima (The University of Tokyo, Japan) |
| 11:55-12:25 | Evaluation of Two Parallel Finite Element
Implementations of the Time-Dependent Advection Diffusion Problem: GPU
versus Cluster Considering Time and Energy Consumption Alberto F. De Souza (Universidade Federal do Espirito Santo, Brazil), Lucas Veronese (Universidade Federal do Espirito Santo, Brazil), Leonardo M. Lima (Instituto Federal de Educacao, Ciencia e Tecnologia do Espirito Santo, Brazil), Claudine Badue (Universidade Federal do Espirito Santo, Brazil) and Lucia Catabriga (Universidade Federal do Espirito Santo, Brazil) |
| Session 3-B (Seminar Room, RIKEN AICS): Cloud & Visualization, Chair: Kenji Ono (RIKEN AICS, Japan) | |
| 10:10-10:40 (Seminar Room, RIKEN AICS) |
no presentation |
| 10:45-11:15 | A Service-Oriented Architecture for Scientific
Computing on Cloud Infrastructures German Molto (Universitat Politecnica de Valencia, Spain), Amanda Calatrava (Universitat Politecnica de Valencia, Spain) and Vicente Hernandez (Universitat Politecnica de Valencia, Spain) |
| 11:20-11:50 | WebViz: Collaborative Visualization System for
Large Scale 3D Data Yichen Zhou (University of Minnesota, USA), Cory Ruegg (Gustavus Adolphus College, USA), Robin Weiss (University of Western Australia, Australia), Erik Sevre (Seoul National University, Korea), Wei Jin (University of Minnesota, USA), Michael Knox (University of Minnesota, USA) and David Yuen (University of Minnesota, USA) |
| 11:55-12:25 | Interactive Volume Rendering based on Ray-Casting
for multi-core architectures Alexandre Nery (Universidade Federal do Rio de Janeiro, Brazil), Nadia Nedjah (Universidade Federal do Rio de Janeiro, Brazil), Felipe M. G. Franca (COPPE-UFRJ, Brazil) and Lech Jozwiak (Eindhoven University of Technology, Netherlands) |
| 12:25-13:30 | (Lunch Break) |
| Invited Talk-5 (Auditorium, Kobe U.), Chair: Taisuke Boku (University of Tsukuba, Japan) | |
| 13:30-14:15 (Auditorium, Kobe U.) |
HPC/PF - High Performance Computing Platform: An
Environment that Accelerates Large-Scale Simulations Kenji Ono (RIKEN Advanced Institute for Computational Science (AICS), Japan) |
| 14:15-14:30 | (Break) |
| Poster Session/Briefing (Lobby/Auditorium, Kobe U.), Chair: Osni Marques (Lawrence Berkeley National Laboratory, USA) | |
| 14:30-17:30 (Lobby/Auditorium, Kobe U.) |
(A-01) Large-scale Magnetostatic Domain Decomposition Analysis Using the Minimal Residual Method Hiroshi Kanayama (Kyushu University), Masao Ogino (Nagoya University), Shin-Ichiro Sugimoto (The University of Tokyo, Japan) and Seigo Terada (Kyushu University) |
| (A-02) Parallelized Adaptive Mesh Refinement Particle-In-Cell Scheme with Dynamic Domain Decomposition Yohei Yagi (Kobe University, Japan), Masaharu Matsumoto (Kobe University/JST-CREST, Japan), Masanori Nunami (NIFS, Japan) and Hideyuki Usui (Kobe Univerisity/JST-CREST, Japan) |
|
| (A-03) Development of a Scalable PIC Simulator for Spacecraft-Plasma Interaction Problems Yohei Miyake (Kobe University, Japan), Hiroshi Nakashima (Kyoto University, Japan) and Hideyuki Usui (Kobe University, Japan) |
|
| (A-04) Fast Active Contour Model and Wavelet Transform for Tumor Segmentation in Medical Image Processing Norma Alias (Universiti Teknologi Malaysia, Malaysia), Hanifah Sulaiman (UITM, Malaysia), Rosdiana Shahril (Universiti Teknologi Malaysia, Malaysia), Arsmah Ibrahim (UITM, Malaysia), Hafizah Farhah Saipol (Universiti Teknologi Malaysia, Malaysia) and Asnida Che Abd. Ghani (Universiti Teknologi Malaysia, Malaysia) |
|
| (A-05) An Architecture Concept for the Scalable Simulation of Dendritic Growth Andreas Schafer (University Erlangen-Nuremberg, Germany) and Dietmar Fey (University Erlangen-Nuremberg, Germany) |
|
| (A-06) Parallel Numerical Simulation of Navier-Stokes and transport equations on GPUs Wesley Menenguci (Universidade Federal do Espirito Santo, Brazil), Lucia Catabriga (Universidade Federal do Espirito Santo, Brazil), Alberto De Souza (Universidade Federal do Espirito Santo, Brazil) and Andrea Valli (Universidade Federal do Espirito Santo, Brazil) |
|
| (N-01) An Implementation of Development Support Middleware for Finite Element Method Application Takeshi Kitayama (The University of Tokyo, Japan), Takeshi Takeda (The University of Tokyo, Japan) and Hiroshi Okuda (The University of Tokyo, Japan) |
|
| (N-02) MGCUDA: An easy programming model for CUDA based multiple GPUs platform Cheng Luo (The University of Tokyo, Japan) and Reiji Suda (The University of Tokyo, Japan) |
|
| (N-03) Construction of Approximated Invariant Subspace for a Real Symmetric Definite Generalized Eigenproblem Using a Linear Combination of Resolvents as the Filter Hiroshi Murakami (Tokyo Metropolitan University, Japan) |
|
| (N-04) OpenMP/MPI Implementation of Tile QR Factorization Algorithm on Multi-Core Cluster Tomohiro Suzuki (University of Yamanashi, Japan), Hideki Miyashita (Software Laboratory Inc., Japan) and Hidetomo Nabeshima (University of Yamanashi, Japan) |
|
| (N-05) Parallel Block Gram-Schmidt Orthogonalization with Optimal Block-size Yoichi Matsuo (Keio University, Japan) and Takashi Nodera (Keio University, Japan) |
|
| (N-06) Implementation of ppOpen-AT into OpenFOAM Satoshi Ito (The University of Tokyo/JST-CREST, Japan), Satoshi Ohshima (The University of Tokyo, Japan) and Takahiro Katagiri (The University of Tokyo, Japan) |
|
| (S-01) File Composition Technique for Improving Access Performance of a Number of Small Files Yoshiyuki Ohno (RIKEN AICS, Japan), Atsushi Hori (RIKEN AICS, Japan) and Yutaka Ishikawa (The University of Tokyo/RIKEN AICS, Japan) |
|
| (S-02) Kernel-Level Blocking MPI Atsushi Hori (RIKEN AICS, Japan) and Yutaka Ishikawa (The University of Tokyo/RIKEN AICS, Japan) |
|
| (S-03) Reordering MPI Ranks for Efficient Barrier Collective Communications on a Multi-Dimensional Torus Network Yoshikazu Kamoshida (The University of Tokyo, Japan) |
|
| 15:45-16:15 (Lounge/Auditorium, Kobe U.) |
(Coffee will be served) |
| VECPAR 2012 Banquet | |
| 18:30-20:30(TBA) | (TBA) |
| 08:30-17:00 (Lobby/Auditorium, Kobe U.) |
(Registration) |
| 09:00-14:00 (Lobby/Auditorium, Kobe U.) |
(Industry Exhibitions) |
| Invited Talk-6 (Auditorium, Kobe U.), Chair: Kengo Nakajima (The University of Tokyo, Japan) | |
| 09:00-09:45 (Auditorium, Kobe U.) |
A Theory for Co-Designing Algorithms and
Architectures under Power and Die-Area Constraints Richard Vuduc (Georgia Institute of Technology, USA) |
| 09:45-10:10 (Lounge/Auditorium, Kobe U.) |
(Coffee Break) |
| Session 4-A (Auditorium, Kobe U.):Performance, Chair: Atsushi Hori (RIKEN AICS, Japan) | |
| 10:10-10:40 (Auditorium, Kobe U.) |
Automatic Generation of the HPC Challenge's
Global FFT Benchmark for BlueGene/P Franz Franchetti (Carnegie Mellon University, USA), Yevgen Voronenko (Accuray Inc., USA) and Gheorghe Almasi (IBM Research, USA) |
| 10:45-11:15 | Matrix multiplication on multidimensional torus
networks Edgar Solomonik (University of California, Berkeley, USA) and James Demmel (University of California, Berkeley, USA) |
| 11:20-11:50 | Determining the Optimal Performance Environment
Configuration for Numerical Libraries Bilel Hadri (University of Tennessee, USA), Haihang You (University of Tennessee, USA) and Shirley Moore (University of Tennessee, USA) |
| Session 4-B (Seminar Room, RIKEN AICS): Methods and Tools for Advanced Scientific Computing, Chair: Richard Vuduc (Georgia Institute of Technology, USA) | |
| 10:10-10:40 (Seminar Room, RIKEN AICS) |
High Performance CPU Kernels for Multiphase
Compressible Flows Babak Hejazialhosseini (ETH Zurich, Switzerland), Christian Conti (ETH Zurich, Switzerland), Diego Rossinelli (ETH Zurich, Switzerland) and Petros Koumoutsakos (ETH Zurich, Switzerland) |
| 10:45-11:15 | Efficient Algorithm for Linear Systems Arising in
Solutions of Eigenproblems and its Application to Electronic-Structure
Calculations
Yasunori Futamura (University of Tsukuba, Japan), Tetsuya Sakurai (University of Tsukuba, Japan), Shinnosuke Furuya (The University of Tokyo, Japan) and Jun-Ichi Iwata (The University of Tokyo, Japan) |
| 11:20-11:50 | Control Formats for Unsymmetric and Symmetric
Sparse Matrix-vector Multiplications Takahiro Katagiri (The University of Tokyo, Japan), Takao Sakurai (Hitachi, Ltd., Japan), Mitsuyoshi Igai (Hitachi ULSI Systems Co., Ltd., Japan), Satoshi Ohshima (The University of Tokyo, Japan), Hisayasu Kuroda (Ehime University/The University of Tokyo, Japan), Ken Naono (Hitachi, Ltd., Japan) and Kengo Nakajima (The University of Tokyo, Japan) |
| 11:50-13:00 | (Lunch Break) |
| Invited Talk-7 (Auditorium, Kobe U.), Chair: Yoshio Oyanagi (Kobe University, Japan) | |
| 13:00-13:45 (Auditorium, Kobe U.) |
Lattice QCD - From Quarks to Nuclei - Yoshinobu Kuramashi (University of Tsukuba/RIKEN, Japan) |
| Session 5-A: Algorithms and Data Analysis, Chair: Michel Dayde (ENSEEIHT, France) | |
| 13:55-14:25 (Auditorium, Kobe U.) |
Parallel implementation of Spectral Clustering Sandrine Mouysset (University of Toulouse, IRIT-UPS, France) and Ronan Guivarch (University of Toulouse, INP(ENSEEIHT)-IRIT, France) |
| 14:30-15:00 | An Experimental Study of Global and Local Search
Algorithms in Empirical Performance Tuning Prasanna Balaprakash (Argonne National Laboratory, USA), Stefan Wild (Argonne National Laboratory, USA) and Paul Hovland (Argonne National Laboratory, USA) |
| 15:05-15:35 | A Multi GPU Read Alignment Algorithm with
Model-based Performance Optimization Aleksandr Drozd (Tokyo Institure of Technology, Japan), Naoya Maruyama (Tokyo Institure of Technology, Japan) and Satoshi Matsuoka (Tokyo Institure of Technology, Japan) |
| Session 5-B (Seminar Room, RIKEN AICS): Parallel Iterative Solvers on Multicore Architectures, Chair: Takahiro Katagiri (The University of Tokyo, Japan) | |
| 13:55-14:25 (Seminar Room, RIKEN AICS) |
OpenMP/MPI Hybrid Parallel ILU(k) Preconditioner
for FEM Based on Extended Hierarchical Interface Decomposition for
Multicore Clusters
Masae Hayashi (The University of Tokyo, Japan) and Kengo Nakajima (The University of Tokyo, Japan) |
| 14:30-15:00 | Parallel Smoother Based on Block Red-Black
Ordering for Multigrid Poisson Solver Masatoshi Kawai (Kyoto University, Japan), Takeshi Iwashita (Kyoto University, Japan), Hiroshi Nakashima (Kyoto University, Japan) and Osni Marques (Lawrence Berkeley National Laboratory, USA) |
| 15:05-15:35 | Software Transactional Memory, OpenMP and Pthread
implementations of the Conjugate Gradients Method - a Preliminary
Evaluation Bjorn Rocker (Robert Bosch GmbH, Germany), Martin Schindewolf (Karlsruher Institut of Technology, Germany) and Sven Janko (Karlsruher Institut of Technology, Germany) |
| 15:35-15:45 | (Break) |
| Closing Session (Auditorium, Kobe U.), Chair: Michel Dayde (ENSEEIHT, France) | |
| 15:45-16:00 (Auditorium, Kobe U.) |
Closing Remarks |