 | New HPCF Website HPCF has a new website at http://www.umbc.edu/hpcf. The new site will contain the most up to date information, including resources for our new cluster tara. We will continue to host hpc information on this site, until the new cluster is completely deployed. |
Welcome to UMBC High Performance Computing Facility (HPCF)! This facility is an interdisciplinary, shared campus resource for scientific computing and research on parallel algorithms. More than 20 researchers from more than 10 departments and research centers from all three colleges of UMBC with the support from the administration created this facility in 2008. This page provides information on the systems in HPCF, documents results of research performed using HPCF machines, resources for users on the usage of the machines, and contact information, plus acknowledgements. The information below is supposed to help researchers by informing about available resources on campus and to direct to the most appropriate machine for their purposes. - Matthias K. Gobbert for the user committee
Systems at UMBC
| New HPCF Website This page now is out of date. Please see http://www.umbc.edu/hpcf/systems.html on our new site. |
Research Projects and Results
Projects from departments and research centers across the UMBC campus have used the HPCF. We have a list of projects that used the HPC cluster and there are a number of projects that have used the kali cluster. Additionally, we provide a preprint server here that is open to publications involving scientific computing, particularly those using machines of HPCF. To submit a preprint, see our contact information page.
Resources for Users
hpc.rs.umbc.edu is a distributed-memory cluster with currently 33 compute nodes, each equipped with two dual-core AMD Opteron processors and 13 GB of memory, connected by an InfiniBand network and with an Infiniband-accessible 14 TB parallel file system. The initial purchase in 2008 pooled funds from several researchers with seed funding from UMBC. This machine will be extended over the next three years up to a total of 128 compute nodes with funding partially supported by the National Science Foundation and additional funding from individual researchers. If you are interested in joining the effort, contact the chair of the user committee: Matthias K. Gobbert (gobbert@math.umbc.edu).If you want general information about this cluster, or if you think you may want to start using our cluster, see our page for prospective users. Users of HPC should look at our page for current users. For information about various research projects based on HPC, see our research projects and results page.
If you need significantly more resources than our cluster can provide, you should consider applying for an account on Teragrid – the NSF terascale computing infrastructure. It consists of over a dozen clusters all over the US, numerous web resources and consulting services. It is described by the Teragrid about page as, "an open scientific discovery infrastructure combining leadership class resources at eleven partner sites to create an integrated, persistent computational resource." See our page about getting access to Teragrid for useful Teragrid-related links, an explanation of the process of obtaining a Teragrid account, and tips about how to write your Teragrid proposal.
If you cannot find what you need, see if the answer to your question is on our frequently asked questions page, contact us or visit us during our in-person support hours.
Contact Information
| New HPCF Website This page is no longer maintained. Please see http://www.umbc.edu/hpcf/about/contact.html on our new site. |
Acknowledgments
| New HPCF Website This page now is no longer maintained. Please see http://www.umbc.edu/hpcf/about/acknowledgements.html on our new site. |
The creation of HPCF would not have been possible without the support of many individuals and organizations. The following information highlights some details. Thanks to all for the help!
The initial purchase of the first 32 nodes of hpc in 2008 was funded jointly by seed funding provided by UMBC and by funds from the individual researchers Larrabee Strow (Physics), Markos Georganopoulos (Physics), Lynn Sparling (Physics), Maricel Kann (Biological Sciences), Curtis Menyuk (Computer Science and Electrical Engineering), Charles Eggleton (Mechanical Engineering), and Dan Bailey (Imaging Research Center).
The philosophy of HPCF as an interdisciplinary, shared campus resource was developed in detail in two MRI proposals to the National Science Foundation (NSF). The proposal in 2008 involved 23 researchers from 10 departments and research centers from all three colleges at UMBC, with PI Matthias K. Gobbert (Mathematics and Statistics), co-PIs Lynn Sparling (Physics), Marie desJardins (Computer Science and Electrical Engineering), Penny Rheingans (Computer Science and Electrical Engineering), and Marc Olano (Computer Science and Electrical Engineering), and co-investigators Florian A. Potra (Mathematics and Statistics), Andrei Draganescu (Mathematics and Statistics), John Zweck (Mathematics and Statistics), Nagaraj K. Neerchal (Mathematics and Statistics), Wallace McMillan (Physics), Markos Georganopoulos (Physics), Larrabee Strow (Physics), Stephen J. Freeland (Biological Sciences), Maricel G. Kann (Biological Sciences), Curtis R. Menyuk (Computer Science and Electrical Engineering), Charles D. Eggleton (Mechanical Engineering), Claire Welty (Civil and Environmental Engineering and CUERE), Erle C. Ellis (Geography and Environmental Systems), Jeffrey B. Halverson (Geography and Environmental Systems and JCET), Scott Farrow (Economics), Andrew Tangborn (JCET), Erricos C. Pavlis (JCET), and Stuart S. Schwartz (CUERE). Some additional faculty involved in the 2007 MRI proposal also helped shape the vision, who were Susan K. Gregurick (Chemistry), Daniele Fabris (Chemistry), Jacob Kogan (Mathematics and Statistics), Howard E. Motteler (Physics), Charles Nicholas (Computer Science and Electrical Engineering), and Philip J. Rous (Physics).
The acquisition of hpc is partially supported by the National Science Foundation, whose support we gratefully acknowledge and which requires the following notice: This material is based upon work supported by the National Science Foundation under Grants CNS-0821258 and DMS-0821311. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
