Salishan 2023 Program
Monday, April 24
4:30pm : Registration opens
Welcome/Keynote Address
6:00pm : An Overview of High Performance Computing and Future Requirements | Jack Dongarra, The University of Tennessee, Oak Ridge National Laboratory, and The University of Manchester
7:00pm : Reception and Informal Discussions
Welcome/Keynote Address
6:00pm : An Overview of High Performance Computing and Future Requirements | Jack Dongarra, The University of Tennessee, Oak Ridge National Laboratory, and The University of Manchester
7:00pm : Reception and Informal Discussions
Tuesday, April 25
7:30am : Breakfast
Session 1 - HPC Applications in the Cloud: Gains and Losses
8:30am : High Performance Multiphysics Applications in the Post-Exascale Era | Rob Rieben, Lawrence Livermore National Laboratory
9:00am : HPC (and Cloud) in Europe and at the Max Planck Society | Erwin Laure and Markus Rampp, Max Planck Computing and Data Facility
9:30am : Industrial Cloud Computing at Rolls-Royce: Is HPC in the Cloud a Cadillac or a Lemon? | Todd Simons, Rolls-Royce
10:00am : Break
10:30am : Why High-Performance Computing Platforms Cannot be Eliminated | Anshu Dubey, Argonne National Laboratory
11:00am : Q&A Panel Session
12:00n : Lunch
Session 2 - Disaggregation and HPC: Promise or Myth
1:30pm : The State of the HPC Market and Market Predictions Regarding Disaggregation and HPC/AI | Earl Joseph, Hyperion
2:00pm : Memory-Centric System Architecture | Pankaj Mehra, Elephance
2:30pm : Rediscovery of Near Data Processing Solutions Thanks to CXL Technology | Jin Lim, SK Hynix
3:00pm : Break
3:30pm : Optical I/O Technology to Meet Future Demands of HPC and AI | Mark Wade, AyarLabs
4:00pm : Q&A Panel Session
6:00pm : Working Dinner
: A Menagerie of High Performance Graphics Systems | Pat Hanrahan, Stanford University
Session 1 - HPC Applications in the Cloud: Gains and Losses
8:30am : High Performance Multiphysics Applications in the Post-Exascale Era | Rob Rieben, Lawrence Livermore National Laboratory
9:00am : HPC (and Cloud) in Europe and at the Max Planck Society | Erwin Laure and Markus Rampp, Max Planck Computing and Data Facility
9:30am : Industrial Cloud Computing at Rolls-Royce: Is HPC in the Cloud a Cadillac or a Lemon? | Todd Simons, Rolls-Royce
10:00am : Break
10:30am : Why High-Performance Computing Platforms Cannot be Eliminated | Anshu Dubey, Argonne National Laboratory
11:00am : Q&A Panel Session
12:00n : Lunch
Session 2 - Disaggregation and HPC: Promise or Myth
1:30pm : The State of the HPC Market and Market Predictions Regarding Disaggregation and HPC/AI | Earl Joseph, Hyperion
2:00pm : Memory-Centric System Architecture | Pankaj Mehra, Elephance
2:30pm : Rediscovery of Near Data Processing Solutions Thanks to CXL Technology | Jin Lim, SK Hynix
3:00pm : Break
3:30pm : Optical I/O Technology to Meet Future Demands of HPC and AI | Mark Wade, AyarLabs
4:00pm : Q&A Panel Session
6:00pm : Working Dinner
: A Menagerie of High Performance Graphics Systems | Pat Hanrahan, Stanford University
Wednesday, April 26
7:30am : Breakfast
Session 3 - Is the Cloud for Everyone?
8:30am : Our Advances Towards FugakuNEXT based on 'Myths in HPC' | Satoshi Matsuoka, RIKEN
9:00am : Big Consequences of Little’s Law | Larry Kaplan, HPE
9:30am : Embracing Communication and Unreliability (Or Life Affter a Pandemic) | Rich Vuduc, Georgia Institute of Technology
10:00am : Break
10:30am : The NSF CloudBank Pilot: Progress and Challenges Facilitating Cloud Adoption | Mike Norman, San Diego Supercomputing Center
11:00am : Q&A Panel Session
12:00n : Lunch on your own
Random Access
5:10pm : Generative AI: Can We Train Computers to do Scientific Computing? | John Daly, Laboratory for Physical Sciences
5:20pm : Myths and Opportunities for Chiplets in HPC | John Shalf, Lawrence Berkeley National Laboratory
5:30pm : Fortran is dead - Long live Fortran! | Torsten Hoefler, ETH Zurich
5:40pm : Enabling Performance Portability, Interoperability, and Novel Features via a DOE Owned Toolchain | Johannes Doerfert, Lawrence Livermore National Laboratory
5:50pm : Performance Portability is (sort of) a Lie | Jonah Miller, Los Alamos National Laboratory
6:00pm : A Practical View of Disaggregation Trade-Offs | Dan Ernst, Microsoft Azure
6:10pm : Embracing the Pain of Irregular Parallelism on HPC and AWS-HPC | Roger Pearce, Lawrence Livermore National Laboratory
6:20pm : The Future is Asynchronous | Vivek Sarkar, Georgia Institute of Technology
6:50pm : Memory Coupled Compute | Bob Wisniewski, Samsung
7:00pm : How the Coming Sea Change in Semiconductors is Going to Change HPC | Bob Patti, NHanced Semiconductors
7:10pm : Energy Efficiency: An Application's Perspective | Valerie Taylor, Argonne National Laboratory
7:20pm : Truths, Myths, and Hype about Memory Technologies | Shekhar Borkhar, Qualcomm
7:30pm : Workflows Are the Future, How Do We Measure Progress? | Ian Karlin, NVIDIA
7:40pm : BPHPC: Broadening Participation in High-Performance Computing | Dorian Arnold, Emory University
7:50pm : Data Provenance and Results Explainability of Scientific Workflows | Michela Taufer, University of Tennessee, Knoxville
8:00pm : Student Poster Session / Informal Discussions
Student Posters
: HPC vs. Cloud: A Graph-Based Benchmarking | Lance Fletcher, Texas A&M University
: Lightweight Kernels for Exploring Co-Design on RISC-V Platforms | Nicholas Gordon, University of Pittsburgh
: Identifying Performance Bottlenecks in Scientific Applications with Call Path Querying | Ian Lumsden, University of Tennessee, Knoxville
: Enabling Scalability in the Cloud for Scientific Workflows: An Earth Science Use Case | Paula Olaya, University of Tennessee, Knoxville
: Next Generation Multiphysics Simulation Code Efforts | Melissa Ann Rasmussen, Utah State University
: Improving MPI Memory Safety for Modern Languages | Jacob Tronge, Kent State University
Session 3 - Is the Cloud for Everyone?
8:30am : Our Advances Towards FugakuNEXT based on 'Myths in HPC' | Satoshi Matsuoka, RIKEN
9:00am : Big Consequences of Little’s Law | Larry Kaplan, HPE
9:30am : Embracing Communication and Unreliability (Or Life Affter a Pandemic) | Rich Vuduc, Georgia Institute of Technology
10:00am : Break
10:30am : The NSF CloudBank Pilot: Progress and Challenges Facilitating Cloud Adoption | Mike Norman, San Diego Supercomputing Center
11:00am : Q&A Panel Session
12:00n : Lunch on your own
Random Access
5:10pm : Generative AI: Can We Train Computers to do Scientific Computing? | John Daly, Laboratory for Physical Sciences
5:20pm : Myths and Opportunities for Chiplets in HPC | John Shalf, Lawrence Berkeley National Laboratory
5:30pm : Fortran is dead - Long live Fortran! | Torsten Hoefler, ETH Zurich
5:40pm : Enabling Performance Portability, Interoperability, and Novel Features via a DOE Owned Toolchain | Johannes Doerfert, Lawrence Livermore National Laboratory
5:50pm : Performance Portability is (sort of) a Lie | Jonah Miller, Los Alamos National Laboratory
6:00pm : A Practical View of Disaggregation Trade-Offs | Dan Ernst, Microsoft Azure
6:10pm : Embracing the Pain of Irregular Parallelism on HPC and AWS-HPC | Roger Pearce, Lawrence Livermore National Laboratory
6:20pm : The Future is Asynchronous | Vivek Sarkar, Georgia Institute of Technology
6:50pm : Memory Coupled Compute | Bob Wisniewski, Samsung
7:00pm : How the Coming Sea Change in Semiconductors is Going to Change HPC | Bob Patti, NHanced Semiconductors
7:10pm : Energy Efficiency: An Application's Perspective | Valerie Taylor, Argonne National Laboratory
7:20pm : Truths, Myths, and Hype about Memory Technologies | Shekhar Borkhar, Qualcomm
7:30pm : Workflows Are the Future, How Do We Measure Progress? | Ian Karlin, NVIDIA
7:40pm : BPHPC: Broadening Participation in High-Performance Computing | Dorian Arnold, Emory University
7:50pm : Data Provenance and Results Explainability of Scientific Workflows | Michela Taufer, University of Tennessee, Knoxville
8:00pm : Student Poster Session / Informal Discussions
Student Posters
: HPC vs. Cloud: A Graph-Based Benchmarking | Lance Fletcher, Texas A&M University
: Lightweight Kernels for Exploring Co-Design on RISC-V Platforms | Nicholas Gordon, University of Pittsburgh
: Identifying Performance Bottlenecks in Scientific Applications with Call Path Querying | Ian Lumsden, University of Tennessee, Knoxville
: Enabling Scalability in the Cloud for Scientific Workflows: An Earth Science Use Case | Paula Olaya, University of Tennessee, Knoxville
: Next Generation Multiphysics Simulation Code Efforts | Melissa Ann Rasmussen, Utah State University
: Improving MPI Memory Safety for Modern Languages | Jacob Tronge, Kent State University
Thursday, April 27
7:30am : Breakfast
Session 4 - Opportunities and Challenges in Adopting Cloud Software Technologies for HPC
8:30am : Why You Should Care About Cloud: Broadening the Reach of HPC | Todd Gamblin, Lawrence Livermore National Laboratory
9:00am : Evaluating the Cloud for Large Scale HPC Workloads | Jack Lange, Oak Ridge National Laboratory
9:30am : Computing-as-a-Service Infrastructure for Accelerating Digital Engineering | Kevin Pedretti, Sandia National Laboratories
10:00am : Break
10:30am : Bringing the Cloud to HPC Before Taking HPC to the Cloud | Quincy Wofford, Los Alamos National Laboratory
11:00am : Q&A Panel Session
12:00n : Lunch
Session 5 - Does HPC / Computing Have a Role in Edge?
1:30pm : Computing at NERSC from the Edge | Katie Antypas, National Energy Research Scientific Computing Center
2:00pm : Space-based Edge Computing | Kim Katko, Los Alamos National Laboratory
2:30pm : Harnessing the Edge-HPC Continuum for Science | Manish Parashar, University of Utah and National Science Foundation
3:00pm : Break
3:30pm : Neuromorphic Computing: How the Brain can inspire computing from HPC to the Edge | Suma Cardwell, Sandia National Laboratories
4:00pm : Q&A Panel Session
5:00pm : informal Discussions
Session 4 - Opportunities and Challenges in Adopting Cloud Software Technologies for HPC
8:30am : Why You Should Care About Cloud: Broadening the Reach of HPC | Todd Gamblin, Lawrence Livermore National Laboratory
9:00am : Evaluating the Cloud for Large Scale HPC Workloads | Jack Lange, Oak Ridge National Laboratory
9:30am : Computing-as-a-Service Infrastructure for Accelerating Digital Engineering | Kevin Pedretti, Sandia National Laboratories
10:00am : Break
10:30am : Bringing the Cloud to HPC Before Taking HPC to the Cloud | Quincy Wofford, Los Alamos National Laboratory
11:00am : Q&A Panel Session
12:00n : Lunch
Session 5 - Does HPC / Computing Have a Role in Edge?
1:30pm : Computing at NERSC from the Edge | Katie Antypas, National Energy Research Scientific Computing Center
2:00pm : Space-based Edge Computing | Kim Katko, Los Alamos National Laboratory
2:30pm : Harnessing the Edge-HPC Continuum for Science | Manish Parashar, University of Utah and National Science Foundation
3:00pm : Break
3:30pm : Neuromorphic Computing: How the Brain can inspire computing from HPC to the Edge | Suma Cardwell, Sandia National Laboratories
4:00pm : Q&A Panel Session
5:00pm : informal Discussions