Conference Program

Conference Program as pdf

Facing the Multicore-Challenge III

September 19-21, 2012
Hochschule für Technik, Stuttgart - University of Applied Sciences

Wednesday, September 19, 2012

 09:30h - 10:15h  Registration and Coffee                                                                       
 10:15h - 10:30h  Welcome and Opening
  Rainer Franke / Rector HFT
 10:30h - 11:30h   Invited Talk
  On the Road to Exaflops - Some Reality Checks
  Bernd Winkelsträter, Fujitsu

 11:30h - 12:00h  Paper
  Invasive Computing Paradigms on High Performance Shared Memory Systems
  Michael Bader, Hans-Joachim Bungartz, Martin Schreiber, TU Munich

 12:00h - 12:30h  Paper
  Array-based Reduction Operations for a Parallel Adaptive FEM
  Martina Balg, Jens Lang, Arnd Meyer, Gudula Rünger, TU Chemnitz


  Lunch Break                                                                                                                                                     

 14:00h - 15:00h Invited Talk    
 MPI-3.0: A Response to New Challenges in Hardware and Software                           
 Torsten Höfler, ETH Zurich

 15:00h - 15:30h

 A similarity-Based Analysis Tool for Porting OpenMP Applications
 Wei Ding, Oscar Hernandez, Barbara Chapman, Univ. of Houston

 15:30h - 15:50h Coffee Break
 15:50h - 16:15h Poster Session
 16:15h - 18:00h
 Short Talk Session I
Ingredients for Ultra-scalable Simulation Codes

Alexander Schäfer, Dietmar Fey, University Erlangen-Nuremberg

Utilizing the StarSs Programming Model for Molecular Dynamics Simulations

Christoph Niethammer, Colin W. Glass, Jose Gracia

Data Structures and Algorithms in Unified Parallel C for Molecular Dynamics

Kamran Idrees, Christoph Niethammer, Colin W. Glass, HLRS Stuttgart

Towards High Performance Optimization

Thomas Baumann, Colin Glass, HLRS, Stuttgart

Automatically Generating Performance-portable Finite Element Solvers for Manycore Architectures

Florian Rathgeber, Graham R. Markall, Lawrence Mitchell, Nicolas Loriant, David A. Ham, Paul H. J. Kelly, Imperial College London

Massively Parallel Particle-in-Cell Simulation

Stefan Siegel, University Würzburg

Challenges for Many-Core Simulation and Visualization — Insights from the
MCSimVis Project

Alexandros Panagiotidis, Daniel Kauker, Thomas Ertl

19:00h - 22:00h                         
 Conference Dinner
 Brauhaus Schönbuch, Bolzstr. 10                                                                                                                                                                        

Thursday, September 20, 2012

 08:30h - 09:00h 
 09:00h -  10:00h     
 Invited Talk
 DEEP and EXTOLL, the Scalable Interconnect for the DEEP Booster Architecture
Mondrian Nüssle, Extoll

 10:00h -  10:30h Paper
 An Efficient High Performance Parallelization of a Discontinuous Galerkin Spectral Element Method
Christoph Altmann, Andrea D. Beck, Florian Hindenlang, Marc Staudenmaier, Gregor. J. Gassner, 
 Claus-Dieter- Munz, Univ. of Stuttgart

 10:30h - 11:00h
 Coffee Break
 11:00h - 11:30h
 Reducing the Memory Footprint of Parallel Applications with KSM
 Nathalie Rauschmeyr, Achim Streit, CERN

11:30h - 12:00h Paper
 Recalibrating Fine-grained Locking in Parallel Bucket Hash Tables
Akos Dudas, Sandor Juhasz, Sandor Kolumban, Budapest University

12:00h - 12:30h Paper
 Impact of Variable Privatization on Extracting Synchronization-Free Slices
Marek Palkowski, West Pomeranian Univ. of Technology

 Lunch Break                                                                                                                                                            

  Large-Scale Computing                                                                                                                 
Stefan Andersson, Cray

 15:00h -15:15h
 Short Talk Session II

 Validating Solutions of Linear Systems on Graphics Processors
Alexander Dallmann, Marco Nehmeier, University Würzburg

 15:15h - 15:35h Coffee Break    
 15:35h - 16:00h Poster Session

 17:00h - 18:00h        
 Social Event
 Mercedes Museum Stuttgart                                           

Friday, September 21, 2012

08:30h - 09:00h Registration
 09:00h - 10:00h
 Invited Talk
 Exascale Preparations using Miniapps
 Richard Barrett, Sandia National Labs

 10:00h - 10:30h
 Parallel Collision Queries on the GPU - A Comparative Study of Different CUDA Implementations
 Rainer Erbes, Anja Mantel, Elmar Schömer, Nicola Wolpert, HFT Stuttgart

 10:30h - 11:00h
 AeminiumGPU: An Intelligent Framework for GPU Programming
Alcides Fonseca, Bruno Cabral, Univ. Coimbra

 11:00h - 11:30h
 Coffee Break                                                            
 11:30h - 12:30h

 Next Generation Programming for Intel                                                                        
Michael Klemm,  Intel

 12:30h -            14:00h
 Lunch Break                                                                                                                                                                                                                                                            

 14:00h - 15:00h Tutorial
 Optimizations for Shared Memory using Threadspotter                                                                          
Royd Lüdtke, Rogue Wave

 15:00h - 15:30h Paper
 Parallel k-Means Image Segmentation Using Sort, Scan &Connected Components on a GPU
Michael Backer, Jan Tünnermann, Bärbel Mertsching, University of Paderborn

 15:30h - 16:00h Coffee Break

 16:00h - 16:45h
 Short Talk Session III     

GASPI - A Partitioned Global Address Space Programming Interface
Jan-Philipp Weiß et al., KIT Karlsruhe + partners

On the Development of a Parallel Domain Decomposition Technique for Simulation of Nonlinear Water Waves
Stefan Lemvig Glimberg, Allan P. Engsig-Karup, Technical University of Denmark                                                                                                                 

The Parareal Algorithm for Multi-layered Parallelism
Allan S. Nielsen, Stefan L. Glimberg, Allan P. Engsig-Karup, Jan S. Hesthaven
, Technical University of Denmark

 16:45h - 17:15h
 Wrap-Up, Prizes and Closing

 Overview of Posters                                                                   
Solving High-dimensional Problems on Processors with Integrated GPU
Alexander Heinecke, TU Munich

Pulsar Searching with Many-Cores
Alessio Sclocco, Rob V. van Nieuwpoort, Amsterdam University, Netherlands eScience Center

Scheduling Overheads for Task-based Parallel Programming Models
Mathias Nachtmann, Colin W Glass, Jose Gracia, HLRS Stuttgart

PINstruct – Efficient Memory Access to Data Structures
Rainer Keller, Shiqing Fan, HFT Stuttgart, HLRS Stuttgart

Development of a GPU-accelerated MIKE 21 Solver for Water Wave Dynamics
Peter Edward Aackermann, Peter Juhler Dinesen Pedersen, Technical University of Denmark

GPU-accelerated and CPU SIMD Optimized Monte Carlo Simulations of phi4 Model
Piotr Bialas, Jakub Kowal, Adam Strzelecki, Jagiellonian University Krakow

Portable Codes on Many-Core Architectures
Mhd. Amer Wafai, Colin W. Glass, Christoph Niethammer, HLRS Stuttgart

GASPI - A Partitioned Global Address Space Programming Interface
Jan-Philipp Weiß et al., KIT Karlsruhe + partners

Parallel Fully Adaptive Tsunami Simulations
Michael Bader, Alexander Breuer, Martin Schreiber, TU Munich

Implementation of Stable Skew-Symmetric Matrix Factorization for Fermi GPUs Using CUDA
Neven Krajina, University of Zagreb

The IMData Approach to Accelerate Data Intensive Workloads
Marcus Völp, Nils Asmussen, Hermann Härtig, TU Dresden

PRAgMaTIc - Parallel Anisotropic Adaptive Mesh Toolkit
Georgios Rokos, Gerard Gorman, Imperial College London

This conference's program will be made available via the Conference4me smartphone application. The application provides a digital, mobile copy of the conference agenda and timetables, venue information and more. Download Conference4me for Android or for iPhone, run the application and choose Facing the Multicore-Challenge III conference from the list of conferences available for download. For more information about Conference4me, including QR codes for direct downloads, please see the official Conference4me website (See

See also previous conference programs: 2010  2011