SC Conference

SC Conference
Name	SC Conference
Status	Active
Genre	Scientific and Technical Conference
Frequency	Annual
Venue	Various international convention centers
Country	International
First	1988
Organizer	Association for Computing Machinery (ACM and IEEE Computer Society)
Attendance	10,000–15,000 (typical)

SC Conference

SC Conference is an annual international meeting focused on high performance computing, networking, storage, and analysis. The conference assembles engineers, researchers, vendors, and policy makers from institutions such as National Science Foundation, Lawrence Livermore National Laboratory, Los Alamos National Laboratory, Argonne National Laboratory, and Sandia National Laboratories. It features peer-reviewed technical papers, tutorials, vendor exhibitions, and competitive program elements drawing participants from Massachusetts Institute of Technology, Stanford University, University of California, Berkeley, Carnegie Mellon University, and University of Illinois Urbana-Champaign.

Overview

SC Conference serves as a forum for dissemination of developments in domains represented by ACM and IEEE Computer Society, and it is closely associated with communities active in projects like TOP500, Green500, SPEC benchmarks, MPI implementations, and OpenMP. Typical components include a technical paper program, poster sessions, workshops, Birds of a Feather gatherings, and an industry exhibition featuring vendors such as NVIDIA, Intel, AMD, IBM, and Cray (now part of HPE). The conference often coincides with announcements from consortia like The Linux Foundation, OpenPOWER Foundation, Kubernetes, OpenStack, and research initiatives funded by European Commission and DARPA.

History

SC Conference originated in the late 1980s, evolving from regional supercomputing symposia into an international flagship event supported by organizations including Association for Computing Machinery and IEEE Computer Society. Early iterations featured contributions from pioneers at Oak Ridge National Laboratory, Lawrence Berkeley National Laboratory, and companies like Cray Research and IBM. Milestones included the formalization of the TOP500 list, collaborations with projects such as LINPACK benchmarking, and the emergence of high-level programming models exemplified by MPI and OpenMP. The conference has tracked transitions from vector architectures to massively parallel processing, the rise of accelerators like GPUs popularized by NVIDIA, the adoption of heterogeneous systems from vendors including Intel and AMD, and the maturation of exascale initiatives championed by U.S. Department of Energy and international partners like European Union research programs.

Topics and Themes

Sessions span computational topics tied to projects and institutions such as LAMMPS, GROMACS, Quantum ESPRESSO, Ansys, and ANSYS Fluent, alongside system-level work referencing LINPACK, HPL, ZettaScaler, and tools from GitHub-hosted communities. Research themes often include performance analysis using tools like Valgrind and Perf, scalable I/O and storage solutions referencing technologies from EMC Corporation and Seagate Technology, and networking innovations linked to InfiniBand, Ethernet, Mellanox Technologies, and standards bodies such as IETF and IEEE 802.3. Data-centric talks draw on collaborations with initiatives like HDF Group, Apache Hadoop, Apache Spark, Pandas, and large-scale facilities such as CERN, Human Genome Project, and National Center for Supercomputing Applications.

Organization and Governance

The conference is organized through a steering committee populated by representatives from sponsoring institutions including ACM, IEEE Computer Society, national laboratories like Oak Ridge National Laboratory and Argonne National Laboratory, and industry partners such as HPE and Dell Technologies. Governance includes program chairs, technical program committees, and local arrangements committees that coordinate with host cities like Denver, Atlanta, Portland, Oregon, Dallas, and Salt Lake City. The peer-review process for the technical program follows practices common to ACM SIGCOMM, IEEE INFOCOM, and other flagship venues, with efforts to ensure reproducibility inspired by initiatives at NeurIPS and ICML.

Notable Proceedings and Awards

Proceedings from SC Conference have published influential papers that later impacted projects at NASA, DOE centers, and enterprises such as Google and Facebook (now Meta Platforms). Awards include best paper recognitions, the Gordon Bell Prize affiliated with ACM and IEEE Computer Society, and technology showcase awards judged by panels with members from National Science Foundation, DOE laboratories, and industry R&D divisions like Intel Labs and NVIDIA Research. Past recipients of recognition have included researchers from Columbia University, Princeton University, University of Cambridge, ETH Zurich, and corporate labs such as Microsoft Research and IBM Research.

Attendance and Impact

Attendance typically ranges from thousands to over ten thousand, drawing delegates from academic institutions such as University of Oxford, University of Tokyo, Tsinghua University, Peking University, and research centers like Riken and INRIA. Industrial participation spans startups and multinational firms including Arm, Broadcom, Western Digital, and cloud providers like Amazon Web Services and Google Cloud Platform. The conference has influenced procurement decisions at national laboratories, inspired collaborations among consortia like OpenMP ARB and MPI Forum, and catalyzed standards discussions involving IEEE Standards Association and ISO. Its legacy includes shaping trajectories for exascale programs, driving adoption of accelerator-based computing, and fostering communities that contribute to open-source projects hosted by Apache Software Foundation and Eclipse Foundation.

Category:Conferences