HCW 2023 Program
The thirty-second Heterogeneity in Computing Workshop (HCW) was held at the Hilton St. Petersburg Bayfront Hotel, St. Petersburg, Florida, on May 15, 2023. All times indicated on this page are in the Eastern Daylight Time (UTC-4) time zone.
Session 1: Introductions and Keynote Presentation (8:45-10 am)
Session Chairs: Behrooz Shirazi (Washington State University, US) and Jan Christian Meyer (Norwegian University of Science and Technology, NO)
Peter Kogge, the Ted H. McCourtney Professor of Computer Science and Engineering at the University of Notre Dame, delivered the HCW 2023 keynote.
Title: Heterogeneity and the Problem that Shall Not Be Named
Abstract: Heterogeneity is clearly increasing, especially as “accelerators” burrow deeper and deeper into different parts of an architecture. What is new, however, is a rapid change in not only the number of such heterogeneous cores, but in their characteristics, such as ISA or memory architecture. This talk is focused on the problem of how to construct efficient programs that combine multiple heterogeneous concurrent threads. We focus on the need today to invoke significant software stacks to cross any of these new boundaries. A suggestion is made of using migrating threads as the glue.
Bio: PETER M. KOGGE is the McCourtney Professor of Computer Science and Engineering at the University of Notre Dame, a retired IBM Fellow, and a founder of a startup Emu Solutions, (now Lucata, Inc.). He is also the ND site director for the multi-university Center for Quantum Technologies. In the past he has been acting Chair of the CSE Dept. and Associate Dean for Research for Engineering at Notre Dame. His research interests are in massively parallel computing paradigms and the interaction of non-traditional technology with computing. He holds over 40 patents and is author of several books, including the 2023 “The Zen of Exotic Computing.” In 2008, he led DARPA’s Exascale technology study group, which resulted in a widely referenced report on how the then next 10 years of technology growth would affect exascale computing. Dr. Kogge is an IEEE and AAAS fellow, and has received the Daniel Slotnick best paper award (1994), the IEEE/ACM Seymour Cray award (2012), the IEEE Charles Babbage award (2014), the IEEE Computer Pioneer award (2015), and the Gauss best paper award from the Int. Supercomputing Conf. (2015).
Break (10-10:30 am)
Session 2: Heterogeneous Tasks and Workflows (10:30 am-12 pm)
Session Chair: Anne Benoit (École Normale Supérieure de Lyon, FR)
A Task Based Approach for Co-Scheduling Ensemble Workloads on Heterogeneous Nodes
Alok Kamatar (University of Chicago, US), Ryan Friese (Pacific Northwest National Laboratory, US), Roberto Gioiosa (Pacific Northwest National Laboratory, US)
Cloud Services Enable Efficient AI-Guided Simulation Workflows across Heterogeneous Resources
Logan Ward (Argonne National Laboratory, US), J. Gregory Pauloski (University of Chicago, US), Valerie Hayot-Sasson (University of Chicago, US), Ryan Chard (Argonne National Laboratory, US), Yadu Babuji (University of Chicago, US), Ganesh Sivaraman (Argonne National Laboratory, US), Sutanay Choudhury (Pacific Northwest National Laboratory, US), Kyle Chard (University of Chicago, US), Rajeev Thakur (Argonne National Laboratory, US), Ian Foster (Argonne National Laboratory, US)
Remote Execution of OpenCL and SYCL Applications via rOpenCL
Rui Alves (Instituto Politécnico de Bragança, PT), Jose Rufino (Instituto Politécnico de Bragança, PT)
Lunch break (12-1:30 pm)
Session 3: Performance Studies and Programming (1:30-3 pm)
Session Chair: Emmanuel Jeannot (INRIA, FR)
Evaluating Energy Efficiency of GPUs using Machine Learning Benchmarks
Brett Foster (Worcester Polytechnic Institute, US), Shubbhi Taneja (Worcester Polytechnic Institute, US), Joseph Manzano (Pacific Northwest National Laboratory, US), Kevin Barker (Pacific Northwest National Laboratory, US)
CEDR-API: Productive, Performant Programming of Domain-Specific Embedded Systems
Joshua Mack (The University of Arizona, US), Serhan Gener (The University of Arizona, US), Sahil Hassan (The University of Arizona, US), H. Umut Suluhan (The University of Arizona, US), Ali Akoglu (The University of Arizona, US)
Power-aware Computing with Optane Persistent Memory Modules
Anara Kozhokanova (RWTH Aachen University, DE), Bo Wang (RWTH Aachen University, DE), Christian Terboven (RWTH Aachen University, DE), Matthias Müller (RWTH Aachen University, DE)
Break (3-3:30 pm)
Session 4: Panel (3:30-5 pm)
Single-chip CPU-GPU HPC architectures: Opportunities and Challenges
Panel Moderator: H. J. Siegel (Colorado State University, US)
Panelists: David Bader (New Jersey Institute of Technology, US), Taisuke Boku (University of Tsukuba, JP), Tsung-Wei Huang (University of Utah, US), Keshav Pingali (University of Texas at Austin, US)