UMBC logo
UMBC High Performance Computing Facility
HPCF Acknowledgments

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 purchase of the 35-node cluster 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 extension of the cluster in 2009 became the replacement purchase of the 86-node cluster tara. We thank the following researchers who added to their earlier contribution or joined the effort: Larrabee Strow (Physics), Wallace McMillan (Physics), Markos Georganopoulos (Physics), Raymond Hoff (Physics and JCET), Charles Eggleton (Mechanical Engineering), Ian Thorpe (Chemistry), Bradford Peercy (Mathematics and Statistics), Do-Hwan Park (Mathematics and Statistics), Weining Kang (Mathematics and Statistics), and Ivan Erill (Biological Sciences).

The purchase of 72 nodes with Intel E5-2650v2 Ivy Bridge CPUs to form the cluster maya extended the QDR InfiniBand interconnect and provided state-of-the-art NVIDIA K20 GPUs designed for scientific computing and cutting-edge Intel Phi 5110P accelerators. This purchase was funded by the 2012 MRI grant as well as significant additional funding from Larrabee Strow (Physics). At the same time, a gift from NASA to UMBC -- the first of its kind -- provided 84 additional nodes very similar to tara, connected by a DDR InfiniBand interconnect. At the same time, UMBC extended the storage dramatically, and maya has a total of more than 750 TB of central storaged connected to it. In total, maya had 240 nodes when released to the public in Summer 2014.

The philosophy of HPCF as a community-based, interdisciplinary core facility was developed in detail in four MRI proposals to the National Science Foundation (NSF).

The acquisition of equipment for the UMBC High Performance Computing Facility 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 the MRI grants CNS-0821258 and CNS-1228778 and the SCREMS grant 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.