Pittsburgh Supercomputing Center

Pittsburgh Supercomputing Center

The Pittsburgh Supercomputing Center is a joint computational research facility located in Pittsburgh, Pennsylvania, formed to provide high-performance computing resources and services to academic, industrial, and government researchers. It partners with Carnegie Mellon University, University of Pittsburgh, National Science Foundation, Department of Energy, and other institutions to support computational science, visualization, and data management across disciplines. The center hosts large-scale systems, supports software development, and carries out collaborative projects with entities such as NASA, National Institutes of Health, National Center for Atmospheric Research, and Lawrence Berkeley National Laboratory.

History

The center was established in 1986 through a collaboration among Carnegie Mellon University, University of Pittsburgh, and the National Science Foundation during an era that included initiatives like the Supercomputer Center program and national investments similar to those supporting Argonne National Laboratory and Oak Ridge National Laboratory. Early leadership built on precedents set by projects involving John von Neumann architectures and supercomputing centers such as San Diego Supercomputer Center and Cornell Theory Center. Over time PSC expanded its mission in tandem with milestones at NASA Ames Research Center and research consortia linked to Intel Corporation and Cray Research. Major programmatic shifts paralleled developments like the Human Genome Project, the rise of climate modeling centers such as Geophysical Fluid Dynamics Laboratory, and collaborations with National Institutes of Health programs for biomedical computing. Directors and principal investigators from institutions including Carnegie Mellon University and University of Pittsburgh guided the center through procurement cycles, community resource allocations, and partnerships modeled after TeraGrid and XSEDE initiatives.

Facilities and Systems

PSC operates computational and visualization facilities housed on university campuses and dedicated centers, with hardware generations comparable to systems at Lawrence Livermore National Laboratory and Los Alamos National Laboratory. Its past and present systems include clusters and parallel machines analogous to products from Cray Inc., Hewlett Packard Enterprise, and Dell EMC, and architectures reflecting designs by NVIDIA and AMD. Storage and networking infrastructures interface with regional fiber networks such as Internet2 and National LambdaRail and utilize file systems and software similar to Lustre and GPFS. Visualization facilities draw on technologies pioneered at Electronic Visualization Laboratory at University of Illinois Urbana–Champaign and infrastructure used by the National Center for Supercomputing Applications. Facilities support scientific workflows relevant to projects involving NOAA, EPA, Smithsonian Institution, and multinational collaborations such as those with European Organization for Nuclear Research researchers.

Research and Services

PSC supports computational science across domains including computational chemistry with ties to American Chemical Society projects, bioinformatics work related to National Institutes of Health initiatives, and climate science connected to Intergovernmental Panel on Climate Change assessments. It provides services such as high-performance computing allocation programs modeled after XSEDE, data management practices akin to DataONE, and software development support reflecting practices from OpenMP and MPI communities. Research collaborations include projects with Pittsburgh Supercomputing Center's academic partners on simulations like molecular dynamics used in studies at Scripps Research, astrophysics computations similar to efforts at Space Telescope Science Institute, and materials science projects complementing work at Argonne National Laboratory and Brookhaven National Laboratory. PSC staff have contributed to middleware and cyberinfrastructure endeavors comparable to Globus and CyVerse, and support reproducible research initiatives consistent with standards promoted by National Academies of Sciences, Engineering, and Medicine reports.

Education and Outreach

PSC runs training programs, workshops, and internships that mirror education efforts at Computer Science Department, Carnegie Mellon University and School of Medicine, University of Pittsburgh, engaging students and researchers in areas such as parallel programming taught with references to CUDA and OpenACC. Outreach connects to K–12 initiatives like FIRST Robotics Competition partnerships and public exhibits similar to programs at the Carnegie Museum of Natural History and Carnegie Science Center. PSC collaborates on curriculum development with university centers including Pittsburgh Supercomputing Center partners to support workforce development initiatives parallel to those by National Science Foundation scholarship programs and regional industry alliances led by Pennsylvania Department of Community and Economic Development stakeholders.

Governance and Funding

Governance is provided through joint oversight by Carnegie Mellon University and University of Pittsburgh, with advisory relationships to federal funders such as the National Science Foundation, and programmatic interactions with agencies including Department of Energy and National Institutes of Health. Funding models include competitive grants, cooperative agreements, and partnerships similar to those used by TeraGrid and XSEDE, supplemented by contracts with industry partners like IBM and Microsoft Research. Strategic planning involves coordination with regional consortia, university leadership, and national research infrastructure bodies such as Office of Science and Technology Policy and advisory committees analogous to those serving National Research Council studies.

Category:Supercomputer centers in the United States Category:Research institutes in Pennsylvania