ACM/IEEE Supercomputing Conference

ACM/IEEE Supercomputing Conference
Name	ACM/IEEE Supercomputing Conference
Status	active
Genre	conference
Frequency	annual
Venue	varies
Country	United States (primarily)
First	1988
Organizer	ACM, IEEE

ACM/IEEE Supercomputing Conference The ACM/IEEE Supercomputing Conference is the flagship annual forum for high-performance computing where researchers, engineers, and industry delegates convene to present advances in parallel computing, networking, and storage. Established to accelerate the development of exascale systems and scalable software, the conference attracts participants from national laboratories, technology companies, and academic institutions. It features peer-reviewed papers, keynote presentations, tutorials, workshops, and an extensive exhibition floor that showcases supercomputer systems, accelerators, and software stacks.

History

The conference evolved from early gatherings in the 1980s that brought together stakeholders from DARPA, Lawrence Livermore National Laboratory, Los Alamos National Laboratory, Sandia National Laboratories, and Argonne National Laboratory and later formalized under the joint stewardship of Association for Computing Machinery and Institute of Electrical and Electronics Engineers leadership. Milestones include programmatic shifts following the advent of vector processors at Cray Research, the emergence of cluster computing at Intel-based installations such as those at Oak Ridge National Laboratory, and pivot points when accelerator-driven architectures from NVIDIA and manycore designs influenced workshop agendas. Historical trends at the conference reflect transitions linked to projects like Blue Gene, Roadrunner, Summit, and initiatives funded by agencies such as the U.S. Department of Energy and National Science Foundation.

Organization and Sponsorship

The event is organized by committees drawn from Association for Computing Machinery special interest groups, IEEE Computer Society, and community volunteers representing universities like Massachusetts Institute of Technology, Stanford University, University of California, Berkeley, and University of Illinois Urbana-Champaign. Sponsorship typically includes technology vendors such as Intel, AMD, NVIDIA, IBM, and cloud providers like Amazon Web Services and Microsoft Azure alongside national funding agencies including the European Commission and Japan Science and Technology Agency. Local hosts have included municipal partners in cities such as Denver, Washington, D.C., Seattle, and Denver International Airport-area venues, with logistical support from conference managers and professional societies like ACM SIGARCH and IEEE Technical Committee on Computer Architecture.

Conferences and Events

Annual programs integrate plenary keynotes from figures associated with projects like Exascale Computing Project, demonstrations of prototype systems such as those from TACC and NERSC, panels featuring leaders from Cray Research successor entities, and vendor exhibits showcasing products from HPE and Fujitsu. Satellite events include tutorials led by faculty from Carnegie Mellon University and University of Cambridge, topical workshops on topics associated with Machine Learning applications in HPC led by teams from Google Research and Facebook AI Research, and doctoral consortia involving students from Princeton University and ETH Zurich. Community-led Birds of a Feather sessions, hackathons co-organized with Linux Foundation projects, and poster sessions featuring researchers from Purdue University and University of Texas at Austin round out the schedule.

Technical Program and Tracks

The technical program covers peer-reviewed papers on architectures, algorithms, programming models, and I/O with participation by contributors affiliated with NVIDIA Research, Intel Labs, IBM Research, Los Alamos National Laboratory, and Argonne National Laboratory. Tracks typically include topics tied to accelerating workloads on architectures associated with GPU, FPGA, and manycore processors, performance modeling inspired by work at Sandia National Laboratories and algorithmic advances from groups at California Institute of Technology and University of Cambridge. Specialized sessions address scaling for irregular applications exemplified by collaborations with teams from Oak Ridge National Laboratory and workflows integrating software ecosystems such as OpenMP, MPI, Kokkos, and TensorFlow adaptations for HPC.

Awards and Recognition

The conference confers awards recognizing influential contributions from researchers and practitioners, with past recipients drawn from institutions such as Stanford University, University of Illinois Urbana-Champaign, Massachusetts Institute of Technology, and national laboratories including Lawrence Berkeley National Laboratory. Honors have acknowledged breakthroughs in system design linked to Cray Research-era innovations, algorithmic milestones associated with John von Neumann-inspired numerical methods, and community service aligned with ACM and IEEE governance. Student paper awards, best paper recognitions, and reproducibility badges highlight impactful work from groups at ETH Zurich, Imperial College London, and Tsinghua University.

Impact and Contributions to HPC

The conference has catalyzed collaborations among hardware vendors, national laboratories, and universities that advanced projects like Blue Gene, Aurora, and Frontier, influencing procurement strategies at centers such as National Energy Research Scientific Computing Center and Texas Advanced Computing Center. Proceedings and technical exchanges have driven standards adoption in programming models like MPI and runtime systems used in production at European Centre for Medium-Range Weather Forecasts and climate modeling groups. By convening developers from NVIDIA, Intel, and AMD alongside scientists from NOAA and NASA, the conference has shaped research agendas in exascale resilience, energy-aware scheduling, and co-design processes fundamental to contemporary high-performance computing.

Category:Computer conferences Category:High performance computing