Waseda University ICT and Robotics

Top Global University Project: Waseda Goes Global -A Plan to Build a Worldwide Academic Network
that is Open, Dynamic and Diverse

Waseda University

Event Information

SISA International Workshop will be held on January 18-19, 2017.
Pre-registration is also available in this link.

SISA Workshop

DateJanuary 18-19, 2017
AvenueGreen Computing Systems Research and Development Center, Waseda University
Target participantsall Waseda students, faculty members, and the general public
Participant feeNone
OrganizerWaseda University international collaboration research in MEXT SGU Embodiment Informatics Project and Waseda University Advanced Multicore Research Institute
SponsorshipIEEE Computer Society Japan Chapter
IEEE Computer Society Multicore STC
IEEE Computer Society Dataflow STC
Industry SupporterFujitsu Limited

Motivation of the workshop

High-performance computers and big-data systems are tied inextricably to the broader computing ecosystem and its designs and markets. However, during the past decade the eco-systems (the programming models, the system software stack, the software tools, etc.) of high-performance computing and big data appear to have been significantly diverged. Such divergence also presents serious challenges to computing architecture and hardware eco-system as well. Such situation has raised significant concerns: many felt that further divergence may be detrimental to both. In particular, both HPC and big-data systems are also facing challenges from the innovations in AI and computational machine learning methods that must be efficiently and securely handled by both. Workshop goal To address these concerns we must study the related open R&D problems that are shared by both HPC and Big-Data systems. The technical topics of this workshop have a system focus on the challenges and opportunities from relevant core technology and applications including novel methods from AI and machine learning. The theme goes beyond supercomputing performance issues and includes talks on the challenges and opportunities from demand of global green computing and security needs — in ways that distinguish them from most other scientific instruments. This workshop provides a unique opportunity to discuss open problems and potential solutions together under the same roof that will have far-reaching impact to future HPC and Big-Data systems, and beyond. We recognize that the challenges are real and the stake is high for global community. And it is far beyond the borders of nations and continents, and we should strongly encourage international participation. We also recognize that besides the traditional leadership in the HPC field (US, EU, Japan), the world has witnessed rapid new advance in other regions such as in Asia: China, India, South Korea, Taiwan to name a few. We hope that this workshop will break new ground on international collaborations inspiring future global exchanges and coordination between the emerging centers in Asia and the established centers in US and Europe.

Program

January 18, 2017

8:15Opening
8:30 – 9:30Keynote session: “The Synergy of HPC and Deep Learning” by William J. Dally, NVIDIA & Stanford University, USA
9:30 – 9:45Break
9:45 – 10:30Kimihiko Hirao, RIKEN, Japan
10:30 – 10:55Guangwen Yang, Tsinghua University, China
10:55 – 11:20James Sexton, IBM, USA
11:20 – 11:30Break
11:30 – 12:30Panel I (Chairs: G. Gao/H. Kasahara)
Panel list: W.J. Dally, M. Sato, G.W. Yang, J. Sexton
12:30 – 14:00Lunch
14:00 – 15:00Keynote session: “Cancer Computing: Driving Architectural Convergence of HPC, Big Data and Deep Learning”
by Rick Stevens, Argonne National Laboratory, USA
15:00 – 15:10Break
15:10 – 15:35Mikhail Smelyanskiy, Intel, USA
15:35 – 16:00Frederick Streitz, Lawrence Livermore National Laboratory, USA
16:00 – 16:25Ramaswamy Govindarajan, Indian Institute of Science, India
16:25 – 16:50Hironori Kasahara, Waseda University, Japan
16:50 – 17:00Break
17:00 – 18:00Panel II (Chairs: G. Gao/H. Kasahara)
Panel list: R. Stevens, M. Smelyanskiy, F. Streitz, R. Govindarajan, H. Kasahara

January 19, 2017

8:30 – 9:30Keynote session: “The U.S. D.O.E. Exascale Computing Project – Goals and Challenges”
by Paul Messina, Argonne National Laboratory, USA
9:30 – 9:45Break
9:45 – 10:10Motoaki Saito, PEZY, Japan
10:10 – 10:35Eiji Ishida, MEXT, Japan
10:35 – 11:00Depei Qian, Beihang University, China
11:00 – 11:25Toshiyuki Shimizu, Fujitsu, Japan
11:25 – 11:35Break
11:35 – 12:35Panel III (Chairs: R. Stevens/T. Sterling)
Panel list: P. Messina, M. Saito, E.Ishida, D. Qian, T. Shimizu, T. Schulthess
12:45 – 14:00Lunch
14:00 – 15:00Keynote session: “Runtime System Architecture for Dynamic Graph Processing”
by Thomas Sterling, Indiana University, USA
15:00 – 15:15Break
15:15 – 16:40Thomas Schulthess, Swiss Federal Institute of Technology (ETH), Switzerland
15:40 – 16:05Moriyuki Takamura/Toshiaki Kitamura, Oscar Tech, Japan
16:05 – 16:50Masaru Kitsuregawa, University of Tokyo, Japan
16:50 – 17:00Break
17:00 – 18:00Panel IV (Chairs: G. Gao/H. Kasahara)
Panel list: T. Sterling, M. Kitsuregawa, T. Schulthess, M. Takamura

Accommodations

RIHGA Royal Hotel Tokyo 1-104-19 Totsuka-machi, Shinjuku-ku, Tokyo 169-8613, Japan TEL:+81-(0)3-5285-1121 » Detailed map

Venue

Green Computing Systems Research Organization 27 Waseda-machi, Shinjuku-ku, Tokyo, Japan 162-0042 TEL:+81-(0)3-3203-4369

Transportation between the hotel and the venue

Walk (approximately 10 minutes)
Public transport (approximately 25-30 minutes, cost around 400 yen)
Taxi (approximately 10 minutes, cost around 730 yen)