NACSE - Northwest Alliance for Computational Science and Engineering

High Performance Computing Sites (U.S.)

The purpose of this page is to provide researchers with a list of sites that provide access to high performance computers. Information is provided regarding system resources, eligibility criteria, and means of access. All of the information below was obtained from the web pages of the various sites. For more information, such as deadlines and application procedures, follow the links to the center web pages.

NSF Centers for HPC:

These 4 sites for high performance computing and communications are directly funded by the National Science Foundation.

Site System Resources Eligibility Means of Access
Cornell Theory Center (CTC), Application Information IBM RS/6000 Scalable POWERparallel SP To apply, you must be a researcher
or educator at a US institution. A principal investigator (PI) may
not be a high school, undergraduate
or graduate student; a qualified
advisor must serve in this capacity.
Postdocs are eligible to serve as PI.
Projects receiving peer-reviewed
allocations must be non-proprietary.
Pittsburgh Supercomputing Center (PSC) , Application Information Cray environment: C90, T3D, J90s, YMP/EL;
Cluster of DEC Alpha workstations
Internet or dial-up access.
National Center for Supercomputing Applications (NCSA), Application Information CONVEX Exemplar, Silicon Graphics (SGI) CHALLENGE, SGI POWER CHALLENGE, Thinking Machines Corporation Connection Machine Model 5 (CM-5) Internet or dial-up access.
San Diego Supercomputing Center (SDSC) , Application Information Cray C90, Cray T3D, Intel Paragon, Advanced Scientific Visualization Laboratory (VisLab) Internet or dial-up access.

Together, these supercomputing centers formed the National MetaCenter for Computational Science and Engineering (NMCSE), which is a cooperative effort between them to offer diverse high performance and scalable parallel architectures for scientists and researchers to attack Grand Challenge problems.

Federal Government/Academic HPC Partnerships

These sites exist as partnerships between the federal government, academia, and industry. Most are funded by government departments as national research laboratories and are operated by universities.

Site System Resources Eligibility Means of Access
Army High Performance Computing Research Center (AHPCRC), Application Information Thinking Machines Corp. CM-5, an advanced graphics and visualization laboratory, CRAY Y-MP/C-90/9512, CRAY T3D, CRAY-2/4512 If you are working on AHPCRC research projects you are eligible to apply for high performance computing time. If your research project is currently funded by the Army Research Office, the Air Force Office of Scientific Research, the Office of Naval Research, the Advanced Research Projects Agency, or another agency of the Department of Defense, contact your service/agency approval authority for an application under the DoD HPC Shared Resources Program. If you are not working on AHPCRC projects or DoD projects you are not normally eligible for time. If you choose to apply, include a written justification with your application. Accessible through networks that support TCP/IP.
The Advanced Computing Initiative in Science and Engineering (ACISE) at Lawrence Livermore National Laboratory (LLNL) , Application Information 256-processor CRAY T3D, a 96-PE Meiko CS-2, a DEC 8-CPU AlphaServer 8400 symmetric multiprocessor, and an ONYX Reality Engine 2 Any researcher at LLNL is eligible to request access. Any research staff member, or any faculty member of the U.C. system, including research staff at Lawrence Berkeley National Laboratory (LBNL), may request access. Students wishing to gain access must enlist the cooperation of a faculty, research, LLNL, or LBNL staff member who will act as principal investigator for the proposed project. Internet
Maui High Performance Computing Center (MHPCC) , Application Information IBM RS/6000 Scalable POWERparallel system, Silicon Graphics computers, including Onyx Reality Engine2 systems and Indigo 2 workstations The MHPCC provides services to three main categories of users: (1) Department of Defense Modernization Program participants, (2) Other Department of Defense, Government, and Private Industry customers, and (3) Academic Projects and Education Programs. Review the eligibility criteria before requesting an account. Internet or dial-up access.

CASC Members

This is a partial list of sites in the Coalition of Academic Supercomputing Centers (CASC) which , established in 1989, represents (the bulk of) the national high performance computing and communications infrastructure. Only the members that provide computer access to researchers outside the institution are listed.

Site System Resources Eligibility Means of Access
Alabama Research and Education Network , Alabama, Application Information Cray C94/264 Supercomputing time and related resources are available to Alabama's academic researchers. Nonacademic institutions and industries are encouraged to explore cooperative ventures with AREN. Contact AREN for more information on eligibility. Internet or dial-up access.
Arctic Region Supercomputing Center , Alaska, Application Information CRAY Y-MP, CRAY T3D, visualization laboratory ARSC allocates computing resources to a limited number of academic and non-profit users external to the University of Alaska. Priority is given to research that is related to ARSC and University of Alaska strategic goals. Preference is given to researchers who clearly identify ARSC services that are uniquely suited to their project requirements. Priority is also given to researchers with agency support and the potential of becoming paying customers of ARSC. Internet.
National Energy Research Scientific Computing (NERSC) Center, California, Application Information Cray T3E, Cray J90, Cray C90, SAS (Supercomputing Auxiliary Service) Any researcher who is currently supported as a principal investigator (PI) by the Office of Energy Research (ER) is eligible to apply to use the Cray C-90 and J-932 machines as part of the ER high-performance computing access program.Any current user of the ER high-performance computing access program is eligible to apply for an MPP allocation. Researchers wishing to port codes from a parallel machine at an HPCRC to the T3E may request MPP allocations. Internet.
North Carolina Supercomputing Center at MCNC, North Carolina, Application Information Cray T916 vector supercomputer, Cray T3D, Workstation Cluster including IBM, DEC and SGI machines NCSC provides supercomputing time and related resources to North Carolina's academic researchers, and industrial and commercial customers. Internet or dial-up access.
Oak Ridge National Laboratory, Tennessee, Application Information Intel Paragons : XP/S 5, XP/S 35, XP/S 150 Researchers from government or educational sites may apply. Members of PICS, CCS-affiliated Grand Challenges, and ORNL-approved LDRD/independent research projects are granted accounts, subject to approval by their respective PI's. Other potential users must indicate why their application requires parallel computing. Internet.
Ohio Supercomputer Center, Ohio Application Information CRAY Y-MP8/8128, CRAY T3D, Convex SPP12200/EX-16, IBM POWERparallel System - SP2, SGI Power CHALLENGE, Quantum Chemistry Engine (QCE) A Principal Investigator must have a full-time appointment (typically a tenure track faculty appointment) at a college, a university, or a research organization administered by an Ohio academic institution. Students, post-doctoral fellows, visiting scientists, and others who wish to use the facilities may be authorized users on projects headed by an eligible PI. Internet.
Utah Supercomputing Institute , Utah, Application Information IBM SP2, SGI Power Challenge, Workstation Cluster USI resources are available to qualified faculty, students (under faculty supervision), and researchers from any Utah institution of higher education. Internet.
Concurrent Supercomputing Consortium 512-node Paragon, 512-node Delta, among others If you would like to use the facilities of the Consortium, you must be affiliated with one of these member institutions: Advanced Research Projects Agency; Argonne National Laboratory, Mathematics and Computer Science Division; California Institute of Technology, Center for Advanced Computing Research; Center for Research on Parallel Computation; Intel Corporation's Scalable Systems Division; Jet Propulsion Laboratory, Caltech; NASA's High-Performance Computing and Communications Program; NASA's Office of Space Science; National Science Foundation; Pacific Northwest Laboratory; Purdue University

And here are some more sites ...

Site System Resources Eligibility Means of Access
Massively Parallel Computing Research Laboratory, Sandia National Laboratories, Application Information Intel/Paragon Access to MPCRL supercomputers is provided to researchers studying massively parallel computer systems and applications. Our machines have four constituencies: MPCRL staff, other Sandia staff, members of the NCHPC (National Consortium for High Performance Computing), and external researchers.Accounts for non-Sandians require a Sandia sponsor. Internet.
Mississippi Center for Supercomputing Research, Application Information Cray Y-MP8D/464, Cray J916, SGI Power Challenge The MCSR provides high-performance computing support to the faculty, professional staff, and students at the eight state-supported senior institutions of higher learning in Mississippi and the University of Mississippi Medical Center. Internet.
The Center for Parallel Computing, University of Michigan, Application Information IBM SP2, HP-Convex Exemplar SPP-1000, SGI Power Challenge New users and especially industrial partners are encouraged to apply. Applicants who are not industrial partners are required to have a UofM faculty sponsor. Internet.

NACSE Home Page
Comments and questions to