Program

Overview Monday Tuesday
Convention Hall A, 2F ( Dec. 12 )
9:30 - 10:40 Session 5 (CREST/SPPEXA) Chair:Shinji Shimojo
9:30 - 10:00 "High Level Programming on Parallel FPGA System for Computation-Communication Unified Computing"
Taisuke Boku
abstract
Abstract Coming soon.
10:00 - 10:20 "Development of a Hierarchical Parallel Solver for Nonlinear Eigenvalue Problems"
Tetsuya Sakurai
abstract
Abstract Coming soon.
10:20 - 10:40 "Preconditioned Iterative Solvers in ppOpen-HPC/pK-Open-HPC for the DFG Project ESSEX"
Kengo Nakajima
abstract
Abstract Coming soon.
10:40 - 12:10 Session 6 (Invited Talk 3) Chair:Keiko Takahashi
10:40 - 11:10 "Correctness Checking for Parallel Programming Paradigms – a progress report from the SPPEXA project MYX"
Matthias S. Mueller
abstract
Abstract Coming soon.
11:10 - 11:40 "Performance Engineering for Scalable Sparse Eigensolvers in the DFG Project ESSEX: From basic building blocks to full scale applications"
Gerhard Wellein
abstract
Abstract Coming soon.
11:40 - 12:10 "Algorithmic Developments and Software Engineering for Scalable Sparse Eigensolvers in the DFG Project ESSEX"
Achim Basermann
abstract
Abstract In the German Research Foundation (DFG) project ESSEX (Equipping Sparse Solvers for Exascale), we develop scalable sparse eigensolver libraries for large quantum physics problems. Partners in ESSEX are the Universities of Erlangen, Greifswald, Wuppertal, Tokyo and Tsukuba as well as DLR. The ESSEX Sparse Sover Repository (ESSR) follows a distributed software development strategy using the distributed version control system Git and supports application driven fault tolerance. ESSR includes the kernel library GHOST (General, Hybrid, and Optimized Sparse Toolit) and the  flexible software framework PHIST for implementing iterative methods on HPC systems. PHIST (Pipelined Hybrid Iterative Solver Toolkit) has been developed containing an interface to the existing numerical software framework Trilinos originally. PHIST also includes adapters to basic building block libraries so that high-level algorithm developments can benefit from high-performance kernel implementations, e.g. sparse matrix-vector multiplication kernels. For solving huge sparse eigenproblems on extreme computing systems, we present our algorithmic approaches regarding Jacobi-Davidson methods with preconditioning, FEAST-like methods and Chebyshev filter diagonalization. The former method is applied to determine extreme eigenvalues and -vectors, while the latter two methods are well suited for solving interior eigenproblems. Moreover, we describe the software architecture of our solver toolkit PHIST.
12:10 - 13:30 Lunch Break
13:30 - 14:30 Session 7 (SPPEXA) Chair:Koichi Kumon
13:30 - 13:50 "Development of a Numerical Library based on ADVENTURE and UG4"
Ryuji Shioya
abstract
Abstract Coming soon.
13:50 - 14:10 "Xevolver: expressing performance-awareness as user-defined code transformations"
Hiroyuki Takizawa
abstract
Abstract Coming soon.
14:10 - 14:30 "Toward embedded domain-specific languages for supercomputing"
Shigeru Chiba
abstract
Abstract Coming soon.
14:30 - 16:00 Session 8 (Invited Talk 4) Chair:Yutaka Ishikawa
14:30 - 15:00 "Highly scalable parallel multigrid simulations on adaptive unstructured grid hierarchies"
Sebastian Reiter
abstract
Abstract Coming soon.
15:00 - 15:30 "Multi-Physics Simulation meets Exa-Scale -- Challenges and Solutions"
Miriam Mehl
abstract
Abstract Coming soon.
15:30 - 16:00 "Scalable multigrid methods with ExaStencils"
Matthias Bolten
abstract
Abstract Coming soon.
16:00 - 16:15 Break @ Hall B
16:15 - 16:55 Session 9 (CREST) Chair:Kenji Kono
16:15 - 16:25 "Process-in-Process: Techniques for Practical Address-Space Sharing"
Atsushi Hori
abstract
Abstract Coming soon.
16:25 - 16:35 "System Software for Data-Intensive Science"
Osamu Tatebe
abstract
Abstract Coming soon.
16:35 - 16:45 "Performance Evaluation of PGAS-based Communication Library ACP"
Takeshi Nanri
abstract
Abstract Coming soon.
16:45 - 16:55 "Advanced Computing & Optimization Infrastructure for Extremely Large-Scale Graphs on Post Peta-Scale Supercomputers"
Katsuki Fujisawa
abstract
Abstract Coming soon.
16:55 - 17:00 Closing