Program (Draft)




Sunday, September 18

17:00-21:00 Check-in & Reception (``Seminario'' - Via Frangipane, 4)
19:00-21:00 Welcome Buffet (``Center Canteen'' - Via Frangipane, 2)


Monday, September 19

07:30-08:45 Breakfast (``Center Canteen'')
08:30-09:00 Registration (``Museum Room'' - La Rocca, Via Frangipane, 6)
09:00-09:15 Welcome
09:15-09:50 Introduction to GPU Architecture
  Pratap Pattnaik (IBM Research)
09:50-10:25 From total Enthusiasm to bitter Cynicism - The religious War on GPUs
  Carsten Trinitis (University of Munich)
10:25-10:55 Coffee
10:55-11:30 EXA-Scale Computing in 2018: About the Challenges for all Kinds of Technology!
  Wolfgang Nagel (Technische Universität Dresden)
11:30-12:30 Discussion Session I: Where are GPUs Going? What are the Technological and Business Opportunities and Constraints? (Part A: Target Questions)
  Coordinators: G. Almasi and P. Kelly
13:00-14:00 Lunch (``Center Canteen'')
15:00-15:35 Hacking GPUs: insights into the NVidia architecture
  Gheorghe Almasi (IBM Research)
15:35-16:10 Parallel Application Characterization with Quantitative Metrics
  Henk Sips (Delft University of Technology)
16:10-16:40 Coffee
16:40-17:15 Experiences in Co-Design: Tackling the Challenges of Performance, Power, and Reliability
  Darren Kerbyson (PNNL)
17:15-17:50 What does it take to Achieve Energy-Efficient Computing. An Exploration of the Software Challenge
  Paul Kelly (Imperial College London)
20:00-21:30 Dinner, Ristorante Belvedere, Via Mazzini 7, 47032 Bertinoro (0543 445127)


Tuesday, September 22

07:30-08:45 Breakfast (``Center Canteen'')
09:00-09:35 The GPU Computational Model as a Bridge to Stream-based Reconfigurable HW Design
  Pedro Trancoso (University of Cyprus)
09:35-10:10 From GPGPUs to Many Core Co-Processors - An Exciting Journey
  Alexander Heinecke (University of Munich)
10:10-10:40 Coffee
10:40-11:15 Where do GPUs Fit: An Analysis based on Case Studies
  Jan Treibig (University of Erlangen)
11:15-12:50 Discussion Session I: Where are GPUs Going? What are the Technological and Business Opportunities and Constraints? (Part B: Answers and Directions for Further Research)
  Coordinators: G. Almasi and P. Kelly
13:00-14:00 Lunch (``Center Canteen'')
14:30-22:30 Excursion to Ravenna
  Social Dinner at Grand Hotel Cesenatico, Piazza Andrea Costa 1, 47042 Cesenatico (0547 80012)



Wednesday, September 21

07:30-08:45 Breakfast (``Center Canteen'')
09:15-09:50 GPU Implementations of Irregular Algorithms
  Keshav Pingali (University of Texas, Austin)
09:50-10:25 SIMD - The Next Generation: Irregular and Non-Numerical Computing
  Jose Moreira (IBM Research)
10:25-11:00 OpenCL vs. Cuda: A Programmability Debate
  Ana Lucia Varbanescu (University of Amsterdam )
11:00-11:30 Coffee
11:30-12:05 Thick Control Flows - Imperative Version of Stream Programming
  Ville Leppanen (University of Turku)
12:05-12:40 Programming Heterogeneous, Accelerator-based Multicore Machines: current situation and main challenges
  Samuel Thibault (LaBRI, University Bordeaux)
13:00-14:00 Lunch (``Center Canteen'')
15:00-16:00 Discussion Session II: What Problems should be Addressed to Realize the GPU Potential? (Part A: Target Questions)
  Coordinators: J. Moreira and K. Pingali
16:00-16:30 Coffee
16:30-17:05 SkePU: A Skeleton Programming Framework for GPU-based Systems
  Usman Dastgeer (Linkoping University)
17:05-17:40 Utilising GPUs by the Press of a Compiler Switch? Lessons from the SaC Approach.
  Sven-Bodo Scholz (University of Hertfordshire )
18:00-18:30 Visit to Bertinoro Inter-Faith Museum (La Rocca, Via Frangipane, 6)
20:00-21:30 Dinner, Casina Pontormo, Via dei Cappuccini, 460, Bertinoro (0543 445699)


Thursday, September 22

07:30-08:45 Breakfast (``Center Canteen'')
09:00-09:35 TBA
  Henry Tufo (CSCS-ETH)
09:35-10:10 GPUs in Astrophysics? Project ISAAC and Multi-science with AMR based GAMER Framework
  Hemant Shukla (LBNL, NERSC)
10:10-10:45 psort, yet another Fast Stable External Sorting Software
  Marco Bressan (University of Padova)
10:45-11:15 Coffee
11:15-11:50 Performance, Efficiency, and Programming Challenges in Many-Core/Exascale Computing Architecture: the General Purpose GPU answer and more
  Lorenzo Verdoscia (ICAR-CNR, Napoli)
11:50-12:25 rCUDA: a Tool for Accessing to a Remote GPU for GPU Computing
  Rafael Mayo Gual (University Jaume I )
12:25-13:00 A Generic Stencil Library
  Mauro Bianco (CSCS-ETH)
13:00-14:00 Lunch (``Center Canteen'')
14:30-16:30 Discussion Session II: What Problems should be Addressed to Realize the GPU potential? (Part B: Answers and Directions for Further Research)
  Coordinators: J. Moreira and K. Pingali
  Adjourn
20:00-21:30 Dinner, Locanda della Fortuna, Via Frangipane 1, 47032 Bertinoro (0543 444456).