Technical Sponsors


     IEEE     IEEE CS   

IEEE Washington DC Section            TCSC 

HPCL


 

PROGRAM

September 16th (Wednesday)

8:30 -9:00 - Breakfast & Registration 

9:00 -10:30 – Tutorials #1 (Room B1220)   and #2 (Room 1300) (Part I)

10:30-11:00 - Break

11:00-12:30 - Tutorials #1 (Room B1220) and #2 (Room 1300) (Part II)

12:30- 1:30 - Lunch

1:30-3:00 - Tutorial #3 (Room B1220) (Part I)

3:00-3:30 - Break

3:30-5:00 – Tutorial #3 (Room B1220) (Part II)


September 17th (Thursday) (Room B1220) 

8:00-8:30 Breakfast & Registration

8:30-8:45 Opening Remarks

8:45-9:45 Keynote Talk I, Exascale Interconnect Architecture: Emerging Developments and Opportunities for PGAS, John Shalf, Lawrence Berkeley National Laboratory 

9:45-10:30 Invited Talk I - The Future of PGAS Programming from a Chapel perspective, Brad Chamberlain, Cray Inc.

10:30-11:00 – Break

11:00-12:30 - Research Paper Session I

Session Chair: Lauren Smith, U. S. Government

  • On the Fence: An Offload Approach to Ordering One-Sided Communication, Mario Flajslik and James Dinan (Intel)
  • Caching Puts and Gets in a PGAS Language Runtime, Michael Ferguson (Cray Inc.) and Daniel Buettner (Laboratory for Telecommunication Sciences)
  • Impact of Frequency Scaling on One Sided Remote Memory Accesses, Siddhartha Jana and Barbara Chapman (University of Houston)

12:30- 1:30 – Lunch

1:30-3:10 - Research Paper Session II

Session Chair: Khaled Hamidouche, The Ohio State University

  • An Implementation of OFI libfabric in Support of Multithreaded PGA Solutions, Sung-Eun Choi (Cray Inc.), Howard Pritchard (LANL), James Shimek (Cray Inc.), James Swaro (Cray Inc.), Zachary Tiffany (Cray Inc.) and Benjamin Turrubiates (LANL)
  • Short Paper: Preliminary Implementation of Coarray Fortran Translator Based on Omni XcalableMP, Hidetoshi Iwashita, Masahiro Nakao and Mitsuhisa Sato (RIKEN AICS)
  • Short Paper: Using the Parallel Research Kernels to study PGAS models, Rob Van der Wijngaart (Intel), Srinivas Sridharan (Intel), Abdullah Kayi (Intel), Gabriele Jost (Intel), Jeff Hammond (Intel), Tim Mattson (Intel), and Jacob Nelson (Univ. of Washington)
  • Short Paper: Hierarchical Locality Exploitation using the PGAS Model, Ahmad Anbar, Olivier Serres, Engin Kayraklioglu, Abdel-Hameed Badawy and Tarek El-Ghazawi (The George Washington University)
  • Poster: A Compiler Transformation to Overlap Communication with Dependent Computation, Karthik Murthy and John Mellor-Crummey (Rice University)
  • Poster: Toward a Data-Centric Profiler for PGAS Applications, Hui Zhang and Jeffrey Hollingsworth (University of Maryland – College Park)
  • Poster: Scaling HabaneroUPC++ on Heterogenous Supercomputers, Vivek Kumar (Rice University), Max Grossman (Rice University), Hongzhang Shan (Lawrence Berkeley National Laboratory) and Vivek Sarkar (Rice University)
  • Poster: PySHMEM: A High Productivity OpenSHMEM Interface for Python, Aaron Welch (University of Houston), Pavel Shamis (Oak Ridge National Laboratory), Pengfei Hao (University of Houston) and Barbara Chapman (University of Houston)
  • Poster: A Scalable Integer Sort for Co-design in the Exascale Era, Ulf Hanebutte and Jacob Hemstad (Intel)

3:10-4:00 – Break and Poster Session

4:00-4:45 - Invited Talk II - Resilient and Elastic APGAS, Olivier Tardieu (IBM)

4:45-5:30 - Invited Talk III - Lessons learned from Using MPI-3 as a PGAS Runtime System, Jeff Hammond (Intel)

7:00 Banquet (Sequoia - 3000 K Street NW Washington, DC 20007)


September 18th (Friday) (Room B1220) 

8:00-8:30 Breakfast & Registration

8:30-9:30 - Keynote Talk II, The Role of Global Address/Name Spaces in Extreme Scale Computing and Analytics, Vivek Sarkar, Rice University

9:30-10:15 - Invited Talk IV – Supporting PGAS and future programming models with the Heterogeneous System Architecture (HSA), Mike Chu (AMD)

10:15-11:00 - Invited Talk V – InfiniBand – A View Towards Extreme-Scale Networking Technology, Rich Graham, Mellanox Technologies

11:00-11:30 - Break

11:30-12:30 - Research Paper Session III

Session Chair: Brad Chamberlain, Cray, Inc. 

  • Implementing High-Performance Geometric Multigrid Solver With Naturally-Grained Messages, Hongzhang Shan, Samuel Williams, Yili Zheng, Amir Kamil and Katherine Yelick (Lawrence Berkeley National Laboratory)
  • Evaluation of Extended Features for Fortran Coarrays, Shiyao Ge, Deepak Eachempati, Dounia Khaldi and Barbara Chapman (University of Houston)

12:30-1:30 – Lunch

1:30 – 3:00 – Panel

The Top Challenges for PGAS in Exascale-Era 
Moderator: Tyler A Simon, Laboratory for Physical Sciences 
Panelists:

  • Bill Carlson (IDA)
  • Jeff Hammond (Intel)
  • John Linford (ParaTools)
  • Fabrizio Petrini (IBM)
  • Kathy Yelick (LBNL)

3:00 – 3:15 – Closing Remarks