Morning Session: Applications on Multicore II                    

- Invited Talk -
The Natural Parallelism
Robert Strzodka, MPI Informatik, Saarbrücken, Germany

Lattice-Boltzmann Simulation of the Shallow-Water Equations
with Fluid-Structure Interaction on Multi- and Manycore Processors
Markus Geveler, Technical University of Dortmund, Germany

Applying Classic Feedback Control for Enhancing the Fault-Tolerance
of Parallel Pipeline Workflows on Multi-Core Systems
Tudor Ionescu, University of Stuttgart, Germany

Programming for Manycore - Challenges and Solutions                                                                                
Moderated Discussion

Analyzing Massive Social Networks using Multicore and Multithreaded Architectures
Wednesday, March 17, 20:15h-21:15h
David A. Bader, Georgia Institute of Technology, USA

MareIncognito: A Perspective Towards Exascale
Thursday, March 18, 9:00h -10:00h
Jesus Labarta, Barcelona Supercomputing Centre, Spain

MareIncognito is a cooperative project between IBM and the Barcelona Supercomputing Center (BSC) targeting the design of relevant technologies on the way towards exascale. The initial challenge of the project was to study the potential design of a system based on a next generation of Cell processors. Even so, the approaches pursued are general purpose, applicable to a wide range of accelerator and homogeneous multicores and holistically addressing a large number of components relevant in the design of such systems.
The programming model is probably the most important issue when facing the multicore era. We need to offer support for asynchronous data flow execution and decouple the way source code looks like and the way the program is executed and its operations (tasks) scheduled. In order to ensure a reasonable migration path for programmers the execution model should be exposed to them through a syntactical and semantic structure that is not very far away from current practice. We are developing the StarSs programming model which we think addresses some the challenges of targeting the future heterogeneous / hierarchical multicore systems at the node level. It also integrates nicely into coarser level programming models such as MPI
and what is more important in ways that propagate the asynchronous dataflow execution to the whole application. We are also investigating how some of the features of StarSs can be integrated in OpenMP.
At the architecture level, interconnect and memory subsystem are two key components. We are studying in detail the behavior of current interconnect systems and in particular contention issues. The question is to investigate better ways to use the raw bandwidth that we already have in our systems and can expect to grow in the future. Better understanding of the interactions
between the raw transport mechanisms, the communication protocols and synchronization behavior of applications should lead to avoid an exploding need for bandwidth that is often claimed. The use of the asynchronous execution model that StarSs offers can help in this direction as a very high overlap between communication and computation should be possible. A similar
effect or reducing sensitivity to latency as well as the actual off chip bandwidth required should be supported by the StarSs model.
The talk will present how we target the above issues, with special details on the StarSs programming model and the underlying idea of the project of how tight cooperation between architecture, run time, programming model, resource management and application are needed in order to achieve in the future the exascale performance.

The Natural Parallelism
Friday, March 19, 9:00h-10:00h
Robert Strzodka, MPI Informatik, Saarbruecken, Germany

• Performance Modeling and Multicore-aware Optimization for 3D Parallel Lattice Boltzmann Simulations, Johannes Habich, University of Erlangen-Nuremberg, Germany
  
• A Multicore Implementation of the Lattice Boltzmann Method for Non-uniform Grids, Kostyantyn Kucher, TU Braunschweig, Germany
  
• A Lattice Boltzmann CUDA-GPU-Implementation on Non-uniform Grids, Martin Schönherr, TU Braunschweig, Germany
  
• Survey of the QPACE Architecture, Nils Meyer, University of Regensburg, Germany
  

• A Pipelined, Multicore-aware Approach to Parallel Temporal Blocking of Stencil
  Codes for Shared and Distributed Memory, Markus Wittman, University of Erlangen-Nuremberg, Germany
• Autotuning Parallel Stencil Computations, Mathias Christen, University of Basel, Switzerland
  
• Mixed Precision in Computational Fluid Dynamics - An Error Correcting Approach
  for Solving Linear Systems, Hartwig Anzt, Karlsruhe Institute of Technology, Germany
  
• MPI, OpenMP and CUDA Approaches for Solving Large Sparse Linear Systems of
  Equations, Benedikt Galler, Karlsruhe Institute of Technology, Germany
• Large Sparse Exact Matching Algorithms for Massive Graph Analysis, Madan Sathe, University of Basel, Switzerland
  
• Numerical Simulation in Computational Finance: Option Pricing with Monte Carlo
  Methods, Philipp Werner, Karlsruhe Institute of Technology, Germany
  

• Applying Software Engineering Methods and Tools to Scientific Research
  Projects-ATLAS Project, Hoda Naguib, TU Munich, Germany
  
• Applying Software Engineering Methods and Tools to the SeisSol Project, Yang Li, TU Munich, Germany 
  
• Single Pattern Multi Value LU Decomposition - Basic Ideas and Parallelization, Martin Köhler, TU Chemnitz, Germany
• Efficient Stereo-image-sequence Segmentation on the GPUs, Alexey Abramov, University of Goettingen, Germany
• Moist Planetary Boundary Layer Simulation Using OpenGL and GLSL, Stefan Horn, Leibniz Institute for Tropospheric Research, Leipzig, Germany
  

• C.M.E.S.S. Solving Large Scale Matrix Equations on Mulitcore Processors, Jens Saak, TU Chemnitz, Germany
  
• Efficient Stereo-image-sequence Segmentation on the GPUs, Alexey Abramov, University of Goettingen, Germany
  
• Mixed Precision in Computational Fluid Dynamics - An Error Correcting Approach for Solving Linear Systems, Hartwig Anzt, Karlsruhe Institute of Technology, Germany
• MPI, OpenMP and CUDA Approaches for Solving Large Sparse Linear Systems of Equations, Benedikt Galler, Karlsruhe Institute of Technology, Germany
  
• Numerical Simulation in Computational Finance: Option Pricing with Monte Carlo Methods, Philipp Werner, Karlsruhe Institute of Technology, Germany
  
• The QPACE Network Processor, Thilo Maurer, University of Regensburg, Germany
  
• Implementation of a Sparse Linear System Solver Utilizing Commodity GPU Hardware and Application to Combustion Simulation, Markus Meingast, TU Berlin, Germany