Workshop Program


Opening Remarks (8:00am - 8:20am)

Pavan Balaji, Abhinav Vishnu, Yong Chen [slides]

Session 1 (8:20am - 10:00am): Programming Models and Performance

Session Chair: Taisuke Boku, University of Tsukuba

  • "When and how VOTM can improve performance in contention situations", Kai-Cheung Leung, Yawen Chen and Zhiyi Huang [slides]
  • "Enhancing MapReduce using MPI and an optimized data exchange policy", Hisham Mohamed and Stéphane Marchand-Maillet [slides]
  • "ALCF MPI Benchmarks: Understanding Machine-Specific Communication Behavior", Vitali Morozov, Jiayuan Meng, Venkatram Vishwanath, Jeff Hammond, Kalyan Kumaran and Michael Papka [slides]
  • "On-the-Fly Synchronization Checking for Interactive Programming in XcalableMP", Tatsuya Abe and Mitsuhisa Sato [slides]

Session 2 (10:20am - 12:30pm): Runtime and Storage Systems

Session Chair: Zhiyi Huang, University of Otago

  • "Chunked Extendible Dense Arrays for Scientific Data Storage", Gideon Nimako, Ekow Otoo and Daniel Ohene-Kwofie [slides]
  • "An Analysis of SMP Memory Allocators: MapReduce on Large Shared-Memory Systems", Robert Döbbelin, Thorsten Schuett and Alexander Reinefeld [slides]
  • "A peer-to-peer architecture for supporting dynamic shared libraries in large-scale systems", Matthew G. F. Dosanjh, Patrick G. Bridges, Suzanne M. Kelly and James H. Laros III [slides]
  • "SWAMP+: Enhanced Smith-Waterman Search for Parallel Models", Shannon Steinfadt [slides]
  • "An Out-Of-Core Dataflow Middleware to Reduce the Cost of Large Scale Iterative Solvers", Zheng Zhou, Erik Saule , Hasan Metin Aktulga, Chao Yang, Esmond G. Ng, Pieter Maris, James P. Vary and Umit V. Catalyurek

Session 3 (1:30pm - 3:00pm): Panel: Battle of the Accelerator Stars

Computing accelerators have gained intensive attentions in recent years. This panel invites experts in accelerated computing to discuss the future of computing accelerators with a theme on hardware/software divergence in accelerator computing. Panelists will discuss questions like the role of accelerator in the next decade, which accelerator and programming model are most likely to win the battle, what research challenges are most critical and should be addressed in the coming years. Panelists will answer questions and discuss with the audience too. Please come to join the panel discussions! [slides]

Panel Moderator: Yong Chen, Texas Tech University [slides]

Panelists:

Pavan Balaji, Argonne National Laboratory [slides]

Taisuke Boku, University of Tsukuba, Japan [slides]

Volodymyr Kindratenko, National Center for Supercomputing Applications (NCSA) [slides]

Xipeng Shen, College of William and Mary and MIT [slides]

Kyle Spafford, Oak Ridge National Laboratory [slides]


Session 4 (3:20pm - 5:50pm): Accelerated Computing and Scheduling

Session Chair: Shannon Steinfadt, Los Alamos National Laboratory

  • "Effective Kernel Mapping for OpenCL Applications in Heterogeneous Platforms", Omer Erdil Albayrak, Ismail Akturk and Ozcan Ozturk [slides]
  • "Lost in Translation: Challenges in Automating CUDA-to-OpenCL Translation", Paul Sathre, Wu-Chun Feng and Mark Gardner [slides]
  • "GPU/CPU Work Sharing with Parallel Language XcalableMP-dev for Accelerated Computing", Tetsuya Odajima, Taisuke Boku, Toshihiro Hanawa, Jinpil Lee and Mitsuhisa Sato [slides]
  • "Adaptive Metric-Aware Job Scheduling for Production Supercomputers", Wei Tang, Dongxu Ren, Zhiling Lan and Narayan Desai [slides]
  • "A Comprehensive Performance Comparison of OpenCL and OpenMP for Multi-core CPUs", Jie Shen, Jianbin Fang, Henk Sips and Ana Lucia Varbanescu [slides]
  • "Cross-Platform OpenCL Code and Performance Portability for CPU and GPU Architectures Investigated with a Climate and Weather Physics Model", Han Dong, Dibyajyoti Ghosh, Fahad Zafar and Shujia Zhou [slides]

Concluding Remarks (5:50pm - 6:00pm)

Pavan Balaji, Abhinav Vishnu, Yong Chen




Workshop Photos