HPCF Acknowledgments

HPCF is a core facility of UMBC. HPCF is proud to partner with many organizations at UMBC and gratefully acknowledges their collaboration and support.

 

 

Additionally, 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!

Faculty Contributions

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 168 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 storage connected to it. In total, maya had 324 nodes when released to the public in Summer 2014.

The purchase in 2018 that created the cluster taki with its CPU cluster of 44 nodes with two 18-core Intel Skylake CPUs and 384 GB of memory each, one node with four NVIDIA Tesla V100 GPUs connected by NVLink, and an 8 node Big Data cluster with 48 TB disk space distributed across 12 hard drives each, was entirely funded by the 2017 MRI grant, see below.

In 2021, 10 faculty from 9 departments and research centers in all 3 colleges at UMBC showed their community spirit by contributing funding to buy 17 additional nodes, each with two 24-core CPUs and 192 GB of memory. These are the faculty listed by college and deparment: CNMS: Joe Bennett (Chemistry) 3 nodes, Pengwang Zhai (Physics) 4 nodes, Larrabee Strow (Physics) 1 node, Matthias Gobbert (Mathematics and Statistics) 2 nodes including one funded by Jerimy Polf (UMB School of Medicine); COEIT: Tyler Josephson (CBEE) 3 nodes, Curtis Menyuk (CSEE) 1 node, Chuck Eggleton (Mechanical Engineering) 1 node; CAHSS: Matt Baker (GES) 1 node, Lee Boot (IRC and IS) 1 node.

MRI Proposals

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

The successful proposal “MRI: Acquisition of a Heterogeneous GPU Cluster to Facilitate Deep Learning Research at UMBC” submitted in 2020 involved 18 researchers from the Department of Computer Science and Electrical Engineering, Department of Information Systems, Department of Mathematics and Statistics, Department of Chemical, Biochemical, and Environmental Engineering, and the Division of Information Technology. The grant includes PI Naghmeh Karimi (original PI Hamed Pirsiavash), co-PIs Cynthia Matuszek, Frank Ferraro, and Damian Doyle, and co-investigator Tulai Adali, Tim Finin, Matthias Gobbert, Anupam Joshi, James Lo, Adam Bargteil, David Chapman, James Foulds, Seung-Jun Kim, Daniel Lobo, Dipanjan Pan, Sanjay Purushotham, Jianwu Wang.

The successful proposal “MRI: Acquisition of Cutting-Edge GPU and Phi Nodes for the Interdisciplinary UMBC High Performance Computing Facility” submitted in 2017 involved 51 researchers from 13 academic departments and research centers across the entire campus of UMBC. The PI was Matthias Gobbert; when he joined the National Science Foundation for one year as rotating program director on an IPA agreement, Meilin Yu took over the responsibilities of the PI for the time. The proposal had the following faculty: PI Matthias Gobbert (Mathematics and Statistics), co-PIs Daniel Lobo (Biological Sciences), Marc Olano (Computer Science and Electrical Engineering), Jianwu Wang (Information Systems), Meilin Yu (Mechanical Engineering), and co-investigators Tulay Adali (Computer Science and Electrical Engineering), Kofi Adragni (Mathematics and Statistics), Matthew Baker (Geography and Environmental Systems), Adam Bargteil (Computer Science and Electrical Engineering), Animikh Biswas (Mathematics and Statistics), Lee Boot (Imaging Research Center (IRC) and Visual Arts), Colleen Burge (Marine Biotechnology), Jian Chen (Computer Science and Electrical Engineering), Fow-Sen Choa (Computer Science and Electrical Engineering), Belay Demoz (Joint Center for Earth Systems Technology (JCET)), Marie desJardins (Computer Science and Electrical Engineering), Sergio DeSouza-Machado (Joint Center for Earth Systems Technology (JCET)), Don Engel (Physics), Ivan Erill (Biological Sciences), Matthew Fagan (Geography and Environmental Systems), Douglas Frey (Chemical, Biochemical and Environmental Engineering (CBEE)), Aryya Gangopadhyay (Information Systems), Upal Ghosh (Chemical, Biochemical and Environmental Engineering (CBEE)), Vandana Janeja (Information Systems), Maricel Kann (Biological Sciences), Jason Kestner (Physics), Seung-Jun Kim (Computer Science and Electrical Engineering), Soobum Lee (Mechanical Engineering), James Lo (Mathematics and Statistics), Thomas Mathew (Mathematics and Statistics), Amita Mehta (Joint Center for Earth Systems Technology (JCET)), Curtis Menyuk (Computer Science and Electrical Engineering), Tinoosh Mohsenin (Computer Science and Electrical Engineering), Nagaraj Neerchal (Mathematics and Statistics), Charles Nicholas (Computer Science and Electrical Engineering), Kevin Omland (Biological Sciences), DoHwan Park (Mathematics and Statistics), Bradford Peercy (Mathematics and Statistics), Carlos Romero-Talamas (Mechanical Engineering), Jon Squire (Computer Science and Electrical Engineering), Michelle Starz-Gaiano (Biological Sciences), Larrabee Strow (Physics), Colin Studds (Geography and Environmental Systems), Michael Summers (Chemistry and Biochemistry), Andrew Tangborn (Joint Center for Earth Systems Technology (JCET)), Ian Thorpe (Chemistry and Biochemistry), Claire Welty (Center for Urban Environmental Research and Education and CBEE), Alan Yeakley (Geography and Environmental Systems), Pengwang Zhai (Physics), Zhibo Zhang (Physics), Weidong Zhu (Mechanical Engineering).

The successful proposal “MRI: Acquisition of Hybrid CPU/GPU Nodes for the Interdisciplinary UMBC High Performance Computing Facility” submitted in 2012 involved 30 researchers from 10 departments and research centers across the entire campus of UMBC, with PI Matthias Gobbert (Mathematics and Statistics), co-PIs Ivan Erill (Biological Sciences), Marc Olano (Computer Science and Electrical Engineering), Lynn Sparling (Physics), and Ian Thorpe (Chemistry), and co-investigators Tulay Adali (Computer Science and Electrical Engineering), Dan Bailey (Imaging Research Center), Matthew Baker (Geography and Environmental Systems), Marie desJardins (Computer Science and Electrical Engineering), Andrei Draganescu (Mathematics and Statistics), Charles Eggleton (Mechanical Engineering), Erle Ellis (Geography and Environmental Systems), Douglas Frey (Chemical and Biochemical Engineering), Markos Georganopoulos (Physics), Raymond Hoff (Physics and JCET), Maricel Kann (Biological Sciences), Lasse Lindahl (Biological Sciences), Curtis Menyuk (Computer Science and Electrical Engineering), Susan Minkoff (Mathematics and Statistics), Tinoosh Mohsenin (Computer Science and Electrical Engineering), Nagaraj Neerchal (Mathematics and Statistics), Tim Oates (Computer Science and Electrical Engineering), Kevin Omland (Biological Sciences), Bradford Peercy (Mathematics and Statistics), Penny Rheingans (Computer Science and Electrical Engineering), Stuart Schwartz (CUERE), Larrabee Strow (Physics), Claire Welty (Civil and Environmental Engineering and CUERE), Zhibo Zhang (Physics), John Zweck (Mathematics and Statistics).

The proposal “MRI: Acquisition of an Expansion of the Interdisciplinary UMBC High Performance Computing Facility” submitted in 2011 involved 41 researchers from 15 departments and research centers across the entire campus of UMBC, with PI Matthias Gobbert (Mathematics and Statistics), co-PIs Ivan Erill (Biological Sciences), Marc Olano (Computer Science and Electrical Engineering), Lynn Sparling (Physics), and Ian Thorpe (Chemistry), and co-investigators Tulay Adali (Computer Science and Electrical Engineering), Dan Bailey (Imaging Research Center), Matthew Baker (Geography and Environmental Systems), Jonathan Bell (Mathematics and Statistics), Marie desJardins (Computer Science and Electrical Engineering), Andrei Draganescu (Mathematics and Statistics), Charles Eggleton (Mechanical Engineering), Erle Ellis (Geography and Environmental Systems), Scott Farrow (Economics), Timothy Finin (Computer Science and Electrical Engineering), Douglas Frey (Chemical and Biochemical Engineering), Chris Geddes (Institute of Fluorescence), Markos Georganopoulos (Physics), Raymond Hoff (Physics and JCET), Maricel Kann (Biological Sciences), Gunes Koru (Information Systems), James Lo (Mathematics and Statistics), Wayne Lutters (Information Systems), Ronghui Ma (Mechanical Engineering), Curtis Menyuk (Computer Science and Electrical Engineering), Susan Minkoff (Mathematics and Statistics), Tinoosh Mohsenin (Computer Science and Electrical Engineering), Nagaraj Neerchal (Mathematics and Statistics), Charles Nicholas (Computer Science and Electrical Engineering), Tim Oates (Computer Science and Electrical Engineering), Kevin Omland (Biological Sciences), Bradford Peercy (Mathematics and Statistics), Roy Rada (Information Systems), Muruhan Rathinam (Mathematics and Statistics), Penny Rheingans (Computer Science and Electrical Engineering), Stuart Schwartz (CUERE), Larrabee Strow (Physics), Haijun Su (Mechanical Engineering), Claire Welty (Civil and Environmental Engineering and CUERE), Liang Zhu (Mechanical Engineering), John Zweck (Mathematics and Statistics).

The successful proposal in 2008 “MRI: Acquisition of an Interdisciplinary Facility for High-Performance Computing” 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 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, CNS-1228778, OAC-1726023, CNS-1920079, 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.