SC16 Salt Lake City, UT

Best Practices in Mentoring Undergraduate Research in Supercomputing


Authors: Prof. Nancy Amato (Texas A&M University)

BP Abstract: We present, discuss, and refine best practices on mentoring undergraduate researchers. We define “best” practices as those which 1)encourage student interest in high-performance computing, 2)produce high quality results, and 3)build student interest in the field. While some might cite the breakdown of Dennard scaling as the largest challenge facing the HPC community, scaling parallel computing education is both a more alarming and challenging issue. The intended audience is graduate students, postdocs, faculty, and researchers who are interested in mentoring undergraduates. The session will comprise a presentation on best practices followed by a discussion of topics of interest to the audience.

Long Description: The proposed BOF session addresses a national imperative of attracting domestic students to pursue graduate degrees and careers in research in computing in general, and supercomputing in particular. While the number of undergraduate students choosing to study computer science and related fields has grown astronomically in recent years, domestic enrollments in graduate programs have grown very modestly in comparison. The health of the research pipeline depends on a larger and more diverse set of students choosing to pursue graduate studies. Most graduate programs struggle to recruit domestic graduate students. It is well known that high quality research experiences can be formative for undergraduates choosing their career trajectories. The purpose of this session is to provide current and future researchers with tools and resources for effectively mentoring undergraduate researchers. The session will describe a number of best practices that have been tested and curated by the faculty leadership of the Computing Research Association’s Education Committee (CRA-E), will include short case studies of research projects in the field of supercomputing that have successfully involved undergraduates, and will allow time for participants to share experiences and pose questions to the facilitators. The discussions and questions that arise from this BoF will be used to guide existing CRA-E activities in encouraging undergraduate involvement in research, and may motivate the proposal of new initiatives. Hence, the desired outcome is two-fold: participants will be able to take the lessons learned by the session leaders back to their respective organizations, and feedback received from the community as part of this session will be used to inform future directions of the CRA-E. The session will be run by Professor Nancy Amato (Texas A&M and member of CRA-E) and Max Grossman (PhD candidate working with Vivek Sarkar at Rice University and Graduate Fellow of CRA-E). Both have been energetic and effective mentors of undergraduate researchers in areas related to supercomputing. Similar sessions have been run in the last two years by CRA-E at major research conferences including ICRA, FCRC, ICSE, and others. This session will be customized for the Supercomputing audience.


Birds of a Feather Index