| This interdisciplinary workshop is organized to explore the scientific issues, challenges, and opportunities for supercomputing beyond the scaling limits of Moore's Law, with the ultimate goal of keeping supercomputing at the forefront of computing technologies beyond the physical and conceptual limits of current systems. Continuing progress of supercomputing beyond the scaling limits of Moore's Law is likely to require a comprehensive re-thinking of technologies, ranging from innovative materials and devices, circuits, system architectures, programming systems, system software, and applications.The workshop is designed to foster interdisciplinary dialog across the necessary spectrum of stakeholders: applications, algorithms, software, and hardware. Motivating workshop questions will include the following. "What technologies might prevail in the Post Moore's Era?" "How can applications effectively prepare for these changes through co-design?" "What architectural abstractions should be in place to represent the traditional concepts like hierarchical parallelism, multi-tier data locality, and new concepts like variable precision, approximate solutions, and resource tradeoff directives?" "What programming models might insulate applications from these changes?"Experts from academia, government, and industry in the fields of computational science, mathematics, engineering, and computer science will have the opportunity to participate in the workshop as a presenter, panelist, or audience member. Invited speakers will provide insights and challenges from their disciplinary perspectives, while peer-reviewed position papers on promising ideas will be presented to facilitate community interaction and diversity. Panel sessions will provide opportunities for interactions across disciplines and provocative questions from the audience.Potential Workshop Topics- Technology trends and predictions
- Alternative memory systems including non-volatile memory
- Superconducting electronics
- Alternative device technologies like CNT transistors
- Quantum computing
- Neuromorphic and brain-inspired computing
- Probabilistic and stochastic computing
- Approximate computing
- 2.5/3-D stacking approaches
- Silicon photonics and optics
- Computational science, data intensive, deep learning application drivers
- Tools for or results from codesign of PMES systems
- Performance studies of PMES systems
- Modeling, simulation, emulation of PMES systems
- Programming paradigms for PMES systems
- Techniques for revolutionary improvements in energy efficient or power constrained supercomputing
Workshop Organizers Workshop Co-Chairs- Prof. Satoshi Matsuoka (Tokyo Institute of Technology)
- Dr. Jeffrey S. Vetter (Oak Ridge National Laboratory)
Program Committee- Prof. Keren Bergman (Columbia)
- Prof. Tom Conte (Georgia Tech)
- Dr. Erik Debenedictis (Sandia National Laboratory)
- Prof. Franz Franchetti (Carnegie Mellon University)
- Prof. Koji Inoue (Kyushu University)
- Prof. Takeshi Iwashita (Hokkaido University)
- Prof. Kengo Nakajima (University of Tokyo)
- John Shalf (Lawrence Berkeley National Laboratory)
- Prof. Richard Vuduc (Georgia Tech)
- Prof. Gerhard Wellein (Erlangen)
|
Important Dates- Submission Site Opens: 17 April 2016
- Submission Deadline:
17 June 2016 extended to July 1 AOE - Notification Deadline: 17 August 2016
- Workshop: 14 November 2016
In cooperation with IEEE Computer Society
|
|
|
|
