GPU Day 2015 - The Future of Many-Core Computing in Science

20 May 2015, 08:30 → 21 May 2015, 18:00 Europe/Budapest

Meeting room (Building 18)

5th GPU WORKSHOP - THE FUTURE OF MANY-CORE COMPUTING IN SCIENCE The conference series organized by the Wigner Research Centre for Physics of the Hungarian Academy of Sciences is dedicated to the field of many-core computing in scientific and industrial applications.

BOOK OF ABSTRACTS AVAILABLE HERE.

TOPICS INCLUDE:

• current status and near-future of many-core hardware and software
• many-core computing in physics and other fields of science
• medical application, image processing, computer vision, reconstruction
• industrial applications of many-core computing
• many-core computing in education
• mobile computing in science
• student projects

LIST OF KEY TALKS:

• NVIDIA Products Overview
  Gernot ZIEGLER (NVIDIA)
   
• GCN Architecture, HSA Platform Evolution and the AMD Developer Ecosystem
  Benjamin COQUELLE (AMD)
   
• SYCL: An Abstraction Layer for Leveraging C++ and OpenCL
  Alastair MURRAY (CODEPLAY)
   
• One Can Simply Walk Into Heterogeneous Parallelism
  Alex VOICU (MICROSOFT)

SPECIAL TOPICAL TALKS:

• Methodology for Heterogeneous Programming on OpenCL and Its Applications
  Olga SUDAREVA (NIISI, Russian Academy of Sciences)
   
• Efficient Large Scale Simulation of Stochastic Lattice Models on GPUs
  Jeffrey KELLING (Helmholtz-Zentrum Dresden-Rossendorf)
   
• GPU Computing at ELI-ALPS
  Sándor FARKAS (ELI Attosecond Light Pulse Source)

CONTRIBUTION TALKS:

• Comparison of GPUs and Embedded Vision Coprocessors for Automotive Computer Vision Applications
  Zoltán PROHÁSZKA (Kishonti Ltd.)
   
• PERSEUS: Image and Video Compression in the Age of Massive Parallelism
  Balázs KESZTHELYI (V-NOVA Ltd.)
   
• GPU-based Iterative Computed Tomography Reconstruction
  Zsolt BALOGH (MEDISO Ltd.)
   
• Fisher Information-based Classification on GPUs
  András BENCZÚR (Institute for Computer Science and Control)
   
• QCD on the lattice
  Ferenc PITTLER (Eötvös University)

• A GPU-based program suite for numerical simulations of planetary system formation
  László DOBOS (Eötvös University)

• Optimization possibilities of inhomogeneous cosmological simulations on massively parallel architectures
  László DOBOS (Eötvös University)

• Semi-analytic modelling of galaxy spectra using GPUs
  Dezső Ribli (Eötvös University)

• Hierarchical Bayesian Method with MCMC Integration on GPUs
  János Márk SZALAI-GINDL (Eötvös University)
   
• GPUs in a Global Earth-based Space Weather Monitoring Network
  Dávid KORONCZAY (Eötvös University)
   
• Pattern Generation for Retina Stimulation Experiments
  László SZÉCSI (IIT, Budapest University of Technology and Economics)
   
• Fast Patient-specific Blood Flow Modelling on GPUs
  Gábor ZÁVODSZKY (HDS, Budapest University of Technology and Economics)
   
• Efficient parallel generation of random p-bits for multispin MC simulation
  István BORSOS (Institute for Technical Physics and Materials Science, Centre for Energy Research)
   
• Accelerating the GEANT Particle Transport System with GPUs
  Gábor BÍRÓ (Eötvös University / Wigner RCP)
   
• Gamma Ray Burst Prediction with GPUs
  Gergely DEBRECZENI (Wigner RCP)
   
• Accelerated Monte-Carlo Particle Generators for the LHC
  Gergely Gábor BARNAFÖLDI (Wigner RCP)
   

CALL FOR ABSTRACTS:

Physics research using or relying on many-core (GPU/APU/MIC) or mobile platform computing solutions are welcome to be presented. Companies and industrial partners are welcome to give presentation and set up discussion panels. Students are encouraged to present their works on posters displayed during the whole duration of the conference.

DEADLINES:

• Abstract submission: 2015 may 3
• Decision on accepted presentations: 2015 may 5
  
  Please send your questions to:
  Dániel BERÉNYI
  {berenyi} . {daniel} {at} {wigner} . {mta} . {hu}

ORGANIZERS:
Gergely DEBRECZENI (Wigner RCP)
Máté Ferenc NAGY-EGRI (Wigner RCP)
Dániel BERÉNYI (Wigner RCP)
Gergely Gábor BARNAFÖLDI (Wigner RCP)

Wednesday, 20 May

08:30 → 09:00 Entry to the campus

09:00 → 09:05 Welcome

Purpose of the conference, practical information

Speaker: Dr Gergely DEBRECZENI (Wigner RCP)

09:05 → 09:35 Current status of GPGPU Computing and Graphics

Máté Ferenc Nagy-Egri

Speaker: Máté Ferenc Nagy-Egri (Wigner Research Centre for Physics, Hungarian Academy of Sciences)

09:35 → 09:55 C++11/14 features relevant in GPGPU APIs

Speaker: Jeffrey Kelling (HZDR)

Slides vortrag.pdf

09:55 → 10:35 CUDA: Present and future perspective

Speaker: Gernot ZIEGLER (Helmholtz-Zentrum Dresden-Rossendorf)

Slides CUDA_7_features_webinar.pdf Tesla_Accelerated_Computing_And_CUDA_7.pdf

10:35 → 10:50 Coffee Break

10:50 → 11:35 OpenCL 2.0, Mantle, Vulkan

Speaker: Benjamin Coquelle (AMD)

Slides OCL2.0.pdf

11:35 → 12:20 C++ AMP

Speaker: Alex Voicu (Microsoft)

Slides CPP_AMP_Budapest.pptx Hetero_simplicity_Budapest.pptx

12:20 → 13:05 SYCL: An Abstraction Layer for Leveraging C++ and OpenCL

Speaker: Alastair Murray (CODEPLAY)

Slides nagy.mate@wigner.hu

13:05 → 14:35 Lunch Break

14:35 → 15:15 Methodology for heterogeneous programming on OpenCL and its applications

Speaker: Olga Sudareva (Institute of System Research Russian Academy of Sciences, НИСИИ РАН)

Slides Bogdanov,_Sudareva_-_GPU-DAY-2015-FINAL.pdf

15:15 → 15:45 GPU computing at ELI-ALPS

Speaker: Sándor Farkas (ELI-ALPS)

Slides nagy.mate@wigner.hu

15:45 → 16:05 Comparison of GPUs and Embedded Vision Coprocessors for Automotive Computer Vision Applications

Speaker: Zoltán Prohászka (Kishonti Ltd.)

16:05 → 16:25 PERSEUS: Image and Video Compression in the Age of Massive Parallelism

Speaker: Balázs Keszthelyi (V-NOVA Ltd.)

Slides nagy.mate@wigner.hu

16:25 → 16:45 Fisher Information-based Classification on GPUs

Speaker: András Benczúr (Informatics Laboratory, Institute for Computer Science and Control, Hungarian Academy of Sciences)

Slides nagy.mate@wigner.hu

16:45 → 17:00 Coffee Break

17:00 → 17:20 A GPU-based program suite for numerical simulations of planetary system formation

Speaker: László Dobos (Eötvös University, Budapest)

17:20 → 17:40 Semi-analytic modelling of galaxy spectra using GPUs

Speaker: Dezső Ribli (Eötvös University, Budapest)

Slides gpunap.pdf gpunap.pptx

17:40 → 18:00 Optimization possibilities of inhomogeneous cosmological simulations on massively parallel architectures

Speaker: László Dobos (Eötvös University, Budapest)

Slides gpunap_backreaction.pdf

18:00 → 18:20 GPUs in a global Earth-based space weather monitoring network

Speaker: Dávid Koronczay (Eötvös University, Budapest)

Slides nagy.mate@wigner.hu

18:20 → 18:40 Forecasting Gamma-Ray Bursts with gravitational wave detectors

Speaker: Gergely Debreczeni (Wigner Research Centre for Physics, Hungarian Academy of Sciences)

Slides 2015.05.20_GPUDay.pdf

Thursday, 21 May

08:30 → 09:00 Entry to the Campus

09:00 → 09:20 NIIF HPC services for research and education

Speaker: Tamás Máray (NIIF)

Slides gpuday15_maray_v3.ppt

09:20 → 10:05 Connected Components Revisited on Kepler

Speaker: Gernot Ziegler (NVIDIA)

Slides S5123_Visual_Computing_in_Automotive_by_Gernot_Ziegler.pdf tesla_deck.pdf

10:05 → 10:50 GCN Architecture, HSA Platform Evolution and the AMD Developer Ecosystem

Speaker: Benjamin Coquelle (AMD)

Slides HSA.pdf OpenCLBestPractices.pdf

10:50 → 11:05 Coffee Break

11:05 → 11:50 One Can Simply Walk Into Heterogeneous Parallelism

Speaker: Alex Voicu (Microsoft)

Slides nagy.mate@wigner.hu

11:50 → 12:30 Efficient Large Scale Simulation of Stochastic Lattice Models on GPUs

Speaker: Jeffrey Kelling (Helmholtz-Zentrum Dresden-Rossendorf)

Slides vortrag.pdf

12:30 → 14:00 Lunch Break

14:00 → 14:20 GPU-based Iterative Computed Tomography Reconstruction

Speaker: Zsolt Balogh (Mediso Ltd.)

Slides nagy.mate@wigner.hu

14:20 → 14:40 Fast patient-specific blood flow modelling on GPUs

Speaker: Gábor Závodszky (Department of Hydrodynamic Systems, Budapest University of Technology and Economics)

Slides nagy.mate@wigner.hu

14:40 → 15:00 Pattern Generation for Retina Stimulation Experiments

Speaker: László Szécsi (Department of Control Engineering and Information Technology, Budapest University of Technology and Economics)

15:00 → 15:20 QCD on the Lattice

Speaker: Ferenc Pittler (Department of Theoretical Physics, Eötvös University, Budapest)

Slides nagy.mate@wigner.hu

15:20 → 15:40 Hierarchical Bayesian Method with MCMC Integration on GPUs

Speaker: János Márk Szalai-Gindl (Eötvös University, Budapest)

Slides Janos_Szalai-Gindl_HB_Method_with_MCMC_Int_on_GPUs.pdf

15:40 → 16:00 Efficient parallel generation of random p-bits for multispin MC simulation

Speaker: István Borsos (Institute for Technical Physics and Materials Science, Centre for Energy Research, Hungarian Academy of Sciences)

16:00 → 16:15 Coffee Break

16:15 → 16:35 Accelerated Monte-Carlo Particle Generators for the LHC

Speaker: Dr Gergely Gábor Barnaföldi (Wigner RCP RMI of the Hungarian Academy of Sciences)

Slides GGBarnafoldi_gpu2015_2015.pdf

16:35 → 16:55 Accelerating the GEANT Particle Transport System with GPUs

Speaker: Gábor Bíró (Eötvös University, Budapest / Wigner Research Centre for Physics, Hungarian Academy of Sciences)

16:55 → 17:15 Code-Generation for Differential Equation Solvers

Speaker: Dániel Berényi (Wigner RCP)

Slides DBerenyi_CodeGenerationforParallelDifferentialEquationSolvers.pdf