MRI: Acquisition of the Bartik High-Performance Computing Cluster

Northwest Missouri State University, Maryville MO

Investigators

Abstract

This project, acquiring a computational cluster named Bartik, supports faculty research employing high-performance computing (HPC) needs and undergraduate education within the home departments of all investigators and senior personnel, and creates opportunities for additional faculty, new or existing, to implement computational aspects into research and/or teaching endeavors in this mainly undergraduate serving institution. (The cluster is named after Jean Bartik, a pioneering woman in the field of computer science and programming and a Northwest alumnus.) Large datasets continue to become increasingly common in many studies. Web-based commerce has allowed marketing data to grow exponentially, and breakthroughs in laboratory technology have enabled collection of vast datasets that were previously unattainable. These data can no longer be stored, manipulated, or analyzed on desktop computers, leaving HPC as a main outlet for many studies. NWMSU interacts with the supercomputing network XSEDE, but relatively few projects require its exceptional resources. The computational cluster would support projects outlined in this proposal and allow benchmarking and piloting of code to be run on XSEDE resources. Bartik would service both the research and education needs at NWMSU undergraduate and graduate students, as well as some high school students that are frequently engaged in faculty research. Research experiences for these students are invaluable and often lead to increased success and retention rates in Science, Technology, Engineering and Math (STEM) fields. This cluster would also support a new undergraduate program in Data Sciences that will begin accepting students in the Fall of 2016. Within this course of study, undergraduate students in Mathematics, Computer Science, Business, Geographic Information Systems (GIS), and Biology will be trained to work with big data using coding, statistical, and visualization approaches.