Overview

The workshop will focus on PGAS (Partitioned Global Address Space) programming model and experiences of the application using PGAS, and the implementation of compiler, runtime and applications.

Its goal is to bring together researchers working in the area of PGAS programming models to parallelize applications, and share the experience using PGAS models. The workshop focuses on PGAS programming languages, tools and libraries with the special focus on experience about the applications developed using this programming model. The workshop is open to any languages and libraries to enables PGAS models. The papers presented practical implementations, scalability and parallelization efficiency are foreseen.

The workshop will be held in conjunction with HPC Asia 2018.

Topics

Possible topics include, but are not limited to:

PGAS Languages and library implementations and their evaluation,
Applications development experience with PGAS programming model,
Benchmark suites and performance studies for PGAS programming model,
Optimization techniques in PGAS models.
PGAS programming model for heterogenous systems,
Comparisons between PGAS and other programming models,
New solutions such as the integrations with multi-tasking, the design for a new architecture, etc, …

Paper Submission

Authors are invited to submit technical papers of at most 10 pages in PDF format including figures and references. Papers should be formatted in the ACM Proceedings Style and submitted at EasyChair

Timeline

Submission Deadline (EXTENDED): November 1, 2017
Acceptance Notification: November 24, 2017
Camera-ready Deadline (EXTENDED): December 25, 2017

Program

13:00 - 13:10 : Welcome
Mitsuhisa Sato (RIKEN AICS)

13:10 - 14:00 : Keynote
Takeshi Nanri (Kyushu University)
Proposal of Interface for Runtime Memory Manipulation of Applications via PGAS-based Communication Library slide

14:00 - 14:30 : Recent Experiences in Using MPI-3 RMA in the DASH PGAS Runtime
Joseph Schuchart, Roger Kowalewski and Karl Fuerlinger slide

14:30 - 15:00 : Investigating the Performance and Productivity of DASH Using the Cowichan Problems
Karl Fuerlinger, Roger Kowalewski, Tobias Fuchs and Benedikt Lehmann slide

15:00 - 15:20 : Coffee Break

15:20 - 15:50 : Linkage of XcalableMP and Python languages for high productivity on HPC cluster system - Application to Graph Order/degree Problem
Masahiro Nakao, Hitoshi Murai, Taisuke Boku and Mitsuhisa Sato slide

15:50 - 16:20 : Performance Evaluation for a Hydrodynamics Application in XcalableACC PGAS Language for Accelerated Clusters
Akihiro Tabuchi, Masahiro Nakao, Hitoshi Murai, Taisuke Boku and Mitsuhisa Sato slide

16:20 – 16:50：Towards a Parallel Algebraic Multigrid Solver Using PGAS
Niclas Jansson and Erwin Laure slide

16:50 - 17:00 : Closing
Mitsuhisa Sato (RIKEN AICS)

Committees

Organizing Committee

Mitsuhisa Sato, RIKEN AICS, Japan
Masahiro Nakao, RIKEN AICS, Japan
Jinpil Lee, RIKEN AICS, Japan

Program Committee

Barbara Chapman, Stony Brook University, USA
Costin Iancu, Lawrence Berkeley National Laboratory, USA
Hidetoshi Iweashita, Fujitsu, Japan
Jaejin Lee, Seoul National University, South Korea
Min Xie, National University of Defense Technology, China
Takeshi Nanri, Kyushu University, Japan
Tarek A. El-Ghazawi, The George Washington University, USA
Yasuharu Hayashi, NEC, Japan
Yi LIU, Beihang University, China
Xingjun ZHANG, Xi’an Jiaotong University, China

Contact

E-mail: jinpil.lee “at” riken.jp