International Workshop on Peta-Scale Computing Programming Environment,

Languages and Tools (WPSE 2010)

The T2K (University of Tsukuba, University of Tokyo, Kyoto University) alliance conducts a project for the seamless and highly-productive parallel programming environment for high-performance computing project to realize a new portable efficient and convenient parallel programming environment for various machines from small scale to large-scale PC clusters and the next generation peta-scale supercomputer. This is a four-year research and development project funded by Ministry of Education, Culture, Sports, Science and Technology, JAPAN. In our project, The following four work packages are carried out: (1) Highly Productive Parallel Programming Language, (2) Highly Productive Parallel Script Language (3) Auto-Tuning Numerical Libraries (4) Portable Single Runtime Environment

In this workshop, we will focus on the research topics of Highly Productive Parallel Programming Language and Portable Single Runtime Environment, and discuss new directions, opportunities and ideas that exploit the seamless and highly-productive parallel programming environment for high-performance peat-scale computing.

Date 18th (Thu) February, 2010
Place Lecture Room 4, Campus Plaza Kyoto
http://csvr15.consortium.or.jp/english/window.html
Registration This workshop is organized on invitation basis.
Program
9:00-9:10 Opening
9:10-9:50 Towards a stable Peta/Exa Scale Computing Envionment [slide] Pete Beckman (Argonne National Laboratory)
9:50-11:50 Library and Script Session
Xabclib: An Iterative Solver with a General Auto-tuning Interface “OpenATLib” [slide] Takahiro Katagiri (University of Tokyo), Takao Sakurai (HITACHI Ltd.), Hisayasu Kuroda (Ehime University/University of Tokyo), Ken Naono (HITACHI Ltd.), Kengo Nakajima (University of Tokyo)
Progress Report of Xcrypt: What You Can Do Now with the Parallel Script Language [slide] Hiroshi Nakashima (Kyoto University)
Lazy functional streams for workflow representation and runtime [slide] Matthew Sottile (University of Oregon)
11:50-13:00 Lunch
13:00-14:20 Parallel Languages Session
Preliminary Performance Report on programming in XcalableMP [slide] Mitsuhisa Sato (University of Tsukuba)
Programming Support for Heterogeneous Parallel Systems [slide] Siegfried Benkner (University of Vienna)
14:20-14:30 Break
14:30-16:40 Runtime Session
Xruntime: MPI-Adapter for Portable MPI Computing Environment [slide] Shinji Sumimoto (Fujitsu Lab.), Atsushi Hori, Kazuki Ohta, Yutaka Ishikawa (University of Tokyo)
Xruntime: Parallel File IO Using Ring Communication Topology – From Viewpoint of File Staging – [slide]
Xruntime: Scalable Parallel I/O Systems by Client-Side Optimizations [slide]
Towards a common runtime system for multicore machines, accelerators and clusters [slide] Raymond Namyst (University of Bordreax, INRIA)
Scientific Data Management and I/O – Challenges for Exascale [slide] Alok N. Choudhary (Northwestern University)
16:40-16:50 Break
16:50-17:30 Current Status of Next Generation Supercomputer in Japan [slide] Mitsuo Yokokawa (RIKEN)
17:30-17:40 Closing
This workshop is supported by “Seamless and Highly-productive Parallel Programming Environment for High-performance computing” project funded by Ministry of Education, Culture, Sports, Science and Technology, JAPAN