Manycore and Accelerator-based High-performance Scientific Computing


Agenda


UC Berkeley/LBNL, Berkeley CA, USA

Tutorials  January 24 – January 25, 2011
Workshop   January 26 – January 28, 2011

 

Auditorium Room 310 Sutardja Dai Hall

Wednesday January 26, 2011
8:00 – 8:15 AM Registration / Refreshments
8:15 – 8:30 AM Horst Simon Welcome
Chair – Hemant Shukla
8:30 – 9:30 AM Kathy Yelick Keynote (LBNL)
9:30 – 10:00 AM Hemant Shukla ICCS Overview – Goals and Roadmap
10:00 – 10:30 AM  Break / Refreshments served
10:30 – 11:00 AM David Bailey 12 ways to fool the masses: Update to 2011
11:00 – 11:30 AM Gregory Newman 3D Geophysical Modeling Imaging of the Subsurface on Multi-Core Machines
11:30 – 12:00 PM Alain Bonissent Accelerated Monte Carlo Simulation of the Xray Radiography: Geant 4 revisited for GPU
12:00 – 1:30 PM  Lunch served
Chair – John Shalf
1:30 – 2:00 PM Jeff Hammond Quantum chemistry many-body methods on GPUs and multicore CPUs
2:00 – 2:30 PM Bryan Catanzaro Copperhead
2:30 – 2:50 PM Jihan Kim GPU Monte Carlo Algorithms for Molecules within a Microporous Framework
2:50 – 3:10 PM Filipe Maia CUDA accelerated X-ray Diffractive Imaging
3:10 – 3:30 PM Steve Hartung GPU Application to Real-time Astronomical Image Pipelines
3:30 – 4:00 PM Mahantesh Halappanavar Multithreaded Algorithms for Approximation and Exact Matching in Graphs
4:00 – 4:30 PM Mous Tatarkhanov Scalability and the real world: lessons learned optimizing ATLAS reconstruction and simulation performance on multicore CPUs.
4:30 – 5:30 PM Poster Session / Refreshments served


Thursday January 27, 2011
8:45 – 9:15 AM Refreshments
Chair – John Shalf
9:15 – 10:00 AM Rainer Spurzem Keynote (NAO/CAS)
10:00 – 10:30 AM Lorena Barba The basis and perspectives of an exascale algorithm: our ExaFMM project
10:30 – 11:00 AM  Break / Refreshments served
11:00 – 11:30 AM Melvyn Wright Adaptive Real Time Imaging for Radio Astronomy
11:30 – 12:00 AM Felipe Cruz /
Tsuyoshi Hamada		 Harnessing heterogeneous systems for N-body problems
12:00 – 12:30 PM Horst Simon Future and next steps
12:30 – 1:00 PM  Lunch served
Chair – Rainer Spurzem
1:00 – 1:30 PM Lincoln Greenhill GPU Computing on a New Frontier in Cosmology
1:30 – 2:00 PM Mike Clark Using GPUs for signal correlation in Radio Astronomy
2:00 – 2:30 PM Hsi-Yu Schive Directionally Unsplit Riemann-Solver-Based Hydrodynamic Schemes in Heterogeneous GPU Computing
2:30 – 3:00 PM Adrian Pope Cosmological Simulations on Large, Accelerated Supercomputers
3:00 – 3:30 PM Break / Refreshments served
3:30 – 4:00 PM Didem Unat CUDA-free programming of 3D stencil methods with Mint
4:00 – 4:30 PM David Sanchez /
David A. Yuen		 Mantle Convection on GPU Using the Rayleigh-Benard Paradigm
4:30 – 5:00 PM Jagan Jayaraj Automatically Generating High Performance Source Code for a Many-Core CPU or Accelerator from a Simplified Input Code Expression


Friday January 28, 2011
8:45 – 9:15 AM Refreshments
Chair – Hemant Shukla
9:15 – 10:00 AM Pradeep Dubey Keynote (Intel)
10:00 – 10:30 AM Paulius Micikevicius Derivative Discretization on GPUs
10:30 – 11:00 AM  Break / Refreshments served
11:00 – 11:30 AM Stan Posey NVIDIA Experiences with Porting Large-Scale Engineering Codes to GPUs
11:30 – 12:00 AM Gilad Shainer The Development of Mellanox - NVIDIA GPUDirect over InfiniBand – a New Model for GPU to GPU Communications
12:00 – 12:30 PM Balint Joo Lattice QCD  on GPU Clusters, using the QUDA Library and the Chroma software system
12:30 – 1:00 PM  Lunch served
Chair – Rainer Spurzem
1:00 – 1:30 PM Niels Joubert Liszt: A Domain Specific Language for solving PDEs on unstructured meshes
1:30 – 2:00 PM Yu Shi Redesigning Combustion Modeling Algorithms for the Graphics Processing Unit (GPU)
2:00 – 2:30 PM I-ju Sung Towards Automation of GPU Program Optimization
2:30 – 3:00 PM Aparna Chandramowlishwaran Performance Evaluation of the 48-core Single Chip Cloud Computer
3:00 – 3:30 PM Break / Refreshments served
3:30 – 4:00 PM David A. Yuen Stokes flow and variable Viscoscity
4:00 – 4:30 PM Assefaw Gebremedhin Multithreaded Graph Coloring Algorithms for Scientific Computing on Many-core Architectures
4:30 – 5:00 PM Jim Bosch MultiFit and Shear Measurement for LSST