============================================================ Call for Participation RSDHA 2021: Workshop on Redefining Scalability for Diversely Heterogeneous Architectures Held in conjunction with SC'21, in cooperation with TCHPC St. Louis, MO, USA November 19, 2021 ============================================================ --Call for Participation-- Recent trends toward the end of Dennard scaling and Moore’s law makes the current and upcoming computing systems more specialized and complex, consisting of more complex and heterogeneous architectures in terms of processors, memory hierarchies, on-chip-interconnection networks, storage, etc. This trend materialized in the mobile and embedded market, and it is entering the enterprise, cloud computing, and high performance computing markets. This workshop will provide a forum for researchers and engineers in both HPC domain and embedded/mobile computing domain to gather together and discuss (1) the latest ideas and lessons learned from the previous experience on traditional (i.e., vertical) and heterogeneity-based (i.e., horizontal) scaling in their own domains and (2) possible synergistic approach and interaction between the two domains to find a good balance between programmability and performance portability across diverse ranges of heterogeneous systems from mobile to HPC. --Agenda-- 8:30am - 8:31am CST Opening Remarks 8:31am - 9:10am CST Invited Talk: Preparing for Extreme Heterogeneity in High Performance Computing Presenter: Jeffrey Vetter (Oak Ridge National Laboratory) 9:10am - 10:10am CST Panel Discussion Moderator: Shuaiwen Leon Song Panelists: Bradford Chamberlain, Allen Malony, Shirley Moore, Catherine Schuman, Antonino Tumeo, and Didem Unat 10:10am - 10:30am CST RSDHA Morning Coffee Break 10:30am - 11:10am CST Invited Talk: Performance-Portable Data Layout Optimizations to Reduce Data Movement Presenter: Mary Hall (University of Utah) 11:10am - 12:00pm CST Paper Presentations 11:10am - 11:23am CST Exploration of Multi-accelerator Neural Network Inference in Diversely Heterogeneous Embedded Systems Author/Presenters: Ismet Dagli and Mehmet Belviranli 11:23am - 11:26am CST Comparing LLC-memory Traffic between CPU and GPU Architectures Author/Presenters: Mohammad Alaul Haque Monil, Seyong Lee, Jeffrey Vetter, and Allen Malony 11:26am - 11:39am CST Platform Agnostic Streaming Data Application Performance Models Author/Presenters: Clayton Faber, Tom Plano, Samatha Kodali, Zhili Xiao, Abhishek Dwaraki, Jeremy Buhler, Roger Chamberlain, and Anthony Cabrera 11:39am - 11:42am CST Distributed Training for High Resolution Images: A Domain and Spatial Decomposition Approach Author/Presenters: Aristeidis Tsaris, Jacob Hinkle, Dalton Lunga, and Philipe Dias 11:42am - 11:55am CST ELIXR: Eliminating Computation Redundancy in CNN-Based Video Processing Author/Presenters: Jordan Schmerge, Daniel Mawhirter, Connor Holmes, Jedidiah McClurg, and Bo Wu 11:55am - 11:58am CST Energy Efficient Task Graph Execution Using Compute Unit Masking in GPUs Author/Presenters: Marcus Chow, Kiran Ranganath, Robert Lerias Jr., Mika Shanela Carodan, and Daniel Wong --Workshop Organizers-- - Mehmet E. Belviranli, Colorado School of Mines, USA Email: belviranli@mines.edu - Seyong Lee, Oak Ridge National Laboratory, USA Email: lees2@ornl.gov - Shuaiwen Leon Song, The University of Sydney, Australia Email: shuaiwen.song@sydney.edu.au -- Technical Program Committee-- - Ali Akoglu, Arizona State University (USA) - Vito Giovanni Castellana, Pacific Northwest National Laboratory (USA) - Kei Davis, Los Alamos National Laboratory (USA) - Javier Teruel Garcia, Barcelona Supercomputing Center (Spain) - Jungwon Kim, Oak Ridge National Laboratory (USA) - Martin Kong, University of Oklahoma (USA) - Selcuk Kose, University of Rochester (USA) - Sriram Krishnamoorthy, Pacific Northwest National Laboratory (USA) - Jaejin Lee, Seoul National University (South Korea) - Pak Markthub, NVIDIA (Japan) - Jedidiah McClurg, Colorado School of Mines (USA) - Eun Jung Park, Los Alamos National Laboratory (USA) - Ivy Peng, Lawrence Livermore National Laboratory (USA) - Amit Sabne, Google (USA) - Dingwen Tao, Washington State University (USA) - Devesh Tiwari, Northeastern University (USA) - Daniel Wong, University of California, Riverside (USA) - Bo Wu, Colorado School of Mines (USA) - Panruo Wu, University of Houston (USA) - Liang Zhou, Facebook (USA)